Convolutional Neural Networks¶
Project: Write an Algorithm for a Dog Identification App¶
In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.
The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.
Why We're Here¶
In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).
In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!
The Road Ahead¶
We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.
- Step 0: Import Datasets
- Step 1: Detect Humans
- Step 2: Detect Dogs
- Step 3: Create a CNN to Classify Dog Breeds (from Scratch)
- Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)
- Step 5: Write your Algorithm
- Step 6: Test Your Algorithm
Step 0: Import Datasets¶
Make sure that you've downloaded the required human and dog datasets:
Note: if you are using the Udacity workspace, you DO NOT need to re-download these - they can be found in the /data folder as noted in the cell below.
Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location
/dog_images.Download the human dataset. Unzip the folder and place it in the home directory, at location
/lfw.
Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.
In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.
In [1]:
import numpy as np
from glob import glob
# load filenames for human and dog images
human_files = np.array(glob("/data/lfw/*/*"))
dog_files = np.array(glob("/data/dog_images/*/*/*"))
# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
There are 13233 total human images. There are 8351 total dog images.
In [19]:
xx=cv2.imread(human_files[0])
plt.imshow(xx)
plt.show()
In [6]:
xy=cv2.imread(dog_files[0])
plt.imshow(xy)
plt.show()
Step 1: Detect Humans¶
In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.
OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.
In [4]:
import cv2
import matplotlib.pyplot as plt
%matplotlib inline
# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')
# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# find faces in image
faces = face_cascade.detectMultiScale(gray)
# print number of faces detected in the image
print('Number of faces detected:', len(faces))
# get bounding box for each detected face
for (x,y,w,h) in faces:
# add bounding box to color image
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Number of faces detected: 1
Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.
In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.
Write a Human Face Detector¶
We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.
In [7]:
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
img = cv2.imread(img_path)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray)
return len(faces) > 0
In [8]:
x=face_detector(dog_files[0]) #check
print(x)
False
(IMPLEMENTATION) Assess the Human Face Detector¶
Question 1: Use the code cell below to test the performance of the face_detector function.
- What percentage of the first 100 images in
human_fileshave a detected human face? - What percentage of the first 100 images in
dog_fileshave a detected human face?
Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.
Answer: (You can print out your results and/or write your percentages in this cell)
human count out of 100 is 98
dog count out of 100 is 17
In [ ]:
In [12]:
from tqdm import tqdm
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
#-#-# Do NOT modify the code above this line. #-#-#
## TODO: Test the performance of the face_detector algorithm
## on the images in human_files_short and dog_files_short.
ch=0
cd=0
for x in human_files_short:
if(face_detector(x)== True):
ch=ch+1
print( 'human count out of 100 is ',ch)
for x in dog_files_short:
if(face_detector(x)== True):
cd=cd+1
print('dog count out of 100 is ',cd)
human count out of 100 is 98 dog count out of 100 is 17
We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
In [ ]:
### (Optional)
### TODO: Test performance of anotherface detection algorithm.
### Feel free to use as many code cells as needed.
Step 2: Detect Dogs¶
In this section, we use a pre-trained model to detect dogs in images.
Obtain Pre-trained VGG-16 Model¶
The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.
In [9]:
import torch
import torchvision.models as models
# define VGG16 model
VGG16 = models.vgg16(pretrained=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# move model to GPU if CUDA is available
if use_cuda:
VGG16 = VGG16.cuda()
Downloading: "https://download.pytorch.org/models/vgg16-397923af.pth" to /root/.torch/models/vgg16-397923af.pth 100%|██████████| 553433881/553433881 [00:05<00:00, 102367327.06it/s]
In [10]:
print(VGG16)
VGG(
(features): Sequential(
(0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): ReLU(inplace)
(2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(3): ReLU(inplace)
(4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(6): ReLU(inplace)
(7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(8): ReLU(inplace)
(9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(11): ReLU(inplace)
(12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(13): ReLU(inplace)
(14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(15): ReLU(inplace)
(16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(18): ReLU(inplace)
(19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(20): ReLU(inplace)
(21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(22): ReLU(inplace)
(23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(25): ReLU(inplace)
(26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(27): ReLU(inplace)
(28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(29): ReLU(inplace)
(30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(classifier): Sequential(
(0): Linear(in_features=25088, out_features=4096, bias=True)
(1): ReLU(inplace)
(2): Dropout(p=0.5)
(3): Linear(in_features=4096, out_features=4096, bias=True)
(4): ReLU(inplace)
(5): Dropout(p=0.5)
(6): Linear(in_features=4096, out_features=1000, bias=True)
)
)
Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.
(IMPLEMENTATION) Making Predictions with a Pre-trained Model¶
In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.
Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.
In [11]:
import torchvision
from PIL import Image
import torchvision.transforms as transforms
def VGG16_predict(img_path):
'''
Use pre-trained VGG-16 model to obtain index corresponding to
predicted ImageNet class for image at specified path
Args:
img_path: path to an image
Returns:
Index corresponding to VGG-16 model's prediction
'''
file=img_path
file = Image.open(file).convert('RGB')
## TODO: Complete the function.
## Load and pre-process an image from the given img_path
transform = transforms.Compose([transforms.RandomResizedCrop(224),
transforms.ToTensor(),
torchvision.transforms.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225))])
#img = transform(file).unsqueeze(0)
t = transform(file)
img = torch.unsqueeze(t, 0)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
VGG16.eval()
with torch.no_grad():
out = VGG16(img.to(device))
#print(out)
ps = torch.exp(out)
#print(ps)
top_p, top_class = ps.topk(1, dim=1)
index = top_class.item()
return index # predicted class index
In [12]:
VGG16_predict(dog_files[0])
Out[12]:
243
(IMPLEMENTATION) Write a Dog Detector¶
While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).
Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).
In [31]:
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
## TODO: Complete the function.
index = VGG16_predict(img_path)
return True if index in range(151, 269) else False
(IMPLEMENTATION) Assess the Dog Detector¶
Question 2: Use the code cell below to test the performance of your dog_detector function.
- What percentage of the images in
human_files_shorthave a detected dog? - What percentage of the images in
dog_files_shorthave a detected dog?
Answer:
dog in human file out of 100 is 2
percentage =2%
dog in dog files count out of 100 is 98
percentage=98%
In [36]:
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
ch1=0
cd1=0
for x in human_files_short:
if(dog_detector(x)== True):
ch1=ch1+1
print( 'dog in human file out of 100 is ',ch1)
for x in dog_files_short:
if(dog_detector(x)== True):
cd1=cd1+1
print('dog in dog files count out of 100 is ',cd1)
dog in human file out of 100 is 2 dog in dog files count out of 100 is 98
We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
In [ ]:
### (Optional)
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.
trying Resnet 50¶
In [34]:
RES50 = models.resnet50(pretrained=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# move model to GPU if CUDA is available
if use_cuda:
RES50 = RES50.cuda()
Downloading: "https://download.pytorch.org/models/resnet50-19c8e357.pth" to /root/.torch/models/resnet50-19c8e357.pth 100%|██████████| 102502400/102502400 [00:01<00:00, 88949670.95it/s]
In [35]:
def RESNET50_predict(img_path):
'''
Use pre-trained RESNET-50 model to obtain index corresponding to
predicted ImageNet class for image at specified path
Args:
img_path: path to an image
Returns:
Index corresponding to ResNet-50 model's prediction
'''
## TODO: Complete the function.
## Load and pre-process an image from the given img_path
## Return the *index* of the predicted class for that image
file=img_path
file = Image.open(file).convert('RGB')
## TODO: Complete the function.
## Load and pre-process an image from the given img_path
transform = transforms.Compose([transforms.RandomResizedCrop(224),
transforms.ToTensor(),
torchvision.transforms.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225))])
#img = transform(file).unsqueeze(0)
t = transform(file)
img = torch.unsqueeze(t, 0)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
RES50.eval()
with torch.no_grad():
out = RES50(img.to(device))
ps = torch.exp(out)
top_p, top_class = ps.topk(1, dim=1)
index = top_class.item()
return index # predicted class index
In [ ]:
def dog_detectorres(img_path):
## TODO: Complete the function.
index = RESNET50_predict(img_path)
return True if index in range(151, 269) else False
In [37]:
ch2=0
cd2=0
for x in human_files_short:
if(dog_detector(x)== True):
ch2=ch2+1
print( 'dog in human file out of 100 is ',ch2)
for x in dog_files_short:
if(dog_detector(x)== True):
cd2=cd2+1
print('dog in dog files count out of 100 is ',cd2)
dog in human file out of 100 is 0 dog in dog files count out of 100 is 100
WE see that here pratrained renet 50 has performed better than VGG16¶
Step 3: Create a CNN to Classify Dog Breeds (from Scratch)¶
Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.
We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.
| Brittany | Welsh Springer Spaniel |
|---|---|
 |
 |
It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).
| Curly-Coated Retriever | American Water Spaniel |
|---|---|
 |
 |
Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.
| Yellow Labrador | Chocolate Labrador | Black Labrador |
|---|---|---|
 |
 |
 |
We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.
Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!
(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset¶
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dog_images/train, dog_images/valid, and dog_images/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!
In [20]:
import os
from torchvision import datasets
#human_files = np.array(glob("/data/lfw/*/*"))
### TODO: Write data loaders for training, validation, and test sets
## Specify appropriate transforms, and batch_sizes
train_dir = '/data/dog_images/train'
test_dir = '/data/dog_images/test'
valid_dir = '/data/dog_images/valid'
In [21]:
train_transforms = transforms.Compose([transforms.RandomRotation(10),
transforms.RandomResizedCrop(224),
transforms.RandomHorizontalFlip(),transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406],[0.229, 0.224, 0.225])])
test_transforms = transforms.Compose([transforms.Resize(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406],[0.229, 0.224, 0.225])])
train_data = datasets.ImageFolder(train_dir , transform=train_transforms)
valid_data = datasets.ImageFolder(valid_dir, transform=test_transforms)
test_data = datasets.ImageFolder(test_dir, transform=test_transforms)
trainloader = torch.utils.data.DataLoader(train_data, batch_size=64, shuffle=True)
validloader = torch.utils.data.DataLoader(valid_data, batch_size=64, shuffle=True)
testloader = torch.utils.data.DataLoader(test_data, batch_size=64)
In [22]:
print(f"training examples contain : {len(train_data)}")
print(f"validation examples contain : {len(valid_data)}")
print(f"testing examples contain : {len(test_data)}")
print('----in each batch we have----- ')
print(len(trainloader))
print(len(validloader))
print(len(testloader))
training examples contain : 6680 validation examples contain : 835 testing examples contain : 836 ----in each batch we have----- 105 14 14
Question 3: Describe your chosen procedure for preprocessing the data.
- How does your code resize the images (by cropping, stretching, etc)? What size did you pick for the input tensor, and why?
- Did you decide to augment the dataset? If so, how (through translations, flips, rotations, etc)? If not, why not?
Answer:
I choose size as 224 * 224 for input tensor by resizing and cropping it using transform
- transforms.Resize(256)
- transforms.CenterCrop(224)
Yes, data augmentations can help in increase the training examples on the fly during training and help to increase the performance of architechture due to randomly indtroduced variations of the sampled examples which i have implemented here using transforms
- transforms.RandomRotation(30)
- transforms.RandomResizedCrop(224)
- transforms.RandomHorizontalFlip()
(IMPLEMENTATION) Model Architecture¶
Create a CNN to classify dog breed. Use the template in the code cell below.
In [68]:
import matplotlib.pyplot as plt
%matplotlib inline
images, labels = next(iter(trainloader))
# Checking shape of image
print(f"Image shape : {images.shape}")
print(f"Label shape : {labels.shape}")
class_names = train_data.classes
def imshow(inp, title=None):
"""Imshow for Tensor."""
inp = inp.numpy().transpose((1, 2, 0))
mean = np.array([0.485, 0.456, 0.406])
std = np.array([0.229, 0.224, 0.225])
inp = std * inp + mean
inp = np.clip(inp, 0, 1)
plt.imshow(inp)
if title is not None:
plt.title(title)
plt.pause(0.001)
fig = plt.figure(figsize=(25, 4))
# display 20 images
for idx in np.arange(20):
ax = fig.add_subplot(2, 20/2, idx+1, xticks=[], yticks=[])
imshow(images[idx])
ax.set_title(class_names[(labels[idx]).item()])
print(labels[idx])
print(class_names[(labels[idx]).item()])
Image shape : torch.Size([64, 3, 224, 224]) Label shape : torch.Size([64])
tensor(82) 083.Ibizan_hound
tensor(6) 007.American_foxhound
tensor(124) 125.Portuguese_water_dog
tensor(32) 033.Bouvier_des_flandres
tensor(68) 069.French_bulldog
tensor(126) 127.Silky_terrier
tensor(17) 018.Beauceron
tensor(60) 061.English_cocker_spaniel
tensor(0) 001.Affenpinscher
tensor(93) 094.Komondor
tensor(45) 046.Cavalier_king_charles_spaniel
tensor(6) 007.American_foxhound
tensor(3) 004.Akita
tensor(91) 092.Keeshond
tensor(39) 040.Bulldog
tensor(51) 052.Clumber_spaniel
tensor(22) 023.Bernese_mountain_dog
tensor(69) 070.German_pinscher
tensor(13) 014.Basenji
tensor(70) 071.German_shepherd_dog
In [60]:
print(class_names[0])
001.Affenpinscher
In [69]:
import torch.nn as nn
import torch.nn.functional as F
# define the CNN architecture
class Net(nn.Module):
### TODO: choose an architecture, and complete the class
def __init__(self):
super(Net, self).__init__()
#conv seess 224*224*3 and we have 133 classes at end
self.conv1 = nn.Conv2d(3, 16, 3, padding=1)
self.conv2 = nn.Conv2d(16, 32, 3, padding=1)
self.conv3 = nn.Conv2d(32, 64, 3, padding=1)
# max pooling layer
self.pool = nn.MaxPool2d(2, 2)
# linear layer (64 * 28 * 28 -> 500)
self.fc1 = nn.Linear(64 * 28 * 28, 500)
# linear layer (500 -> 133)
self.fc2 = nn.Linear(500, 133)
# dropout layer (p=0.25)
self.dropout = nn.Dropout(0.5)
def forward(self, x):
# add sequence of convolutional and max pooling layers
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = self.pool(F.relu(self.conv3(x)))
# flatten image input
x = x.view(-1, 64 * 28 * 28)
# add dropout layer
x = self.dropout(x)
# add 1st hidden layer, with relu activation function
x = F.relu((self.fc1(x)))
# add dropout layer
x = self.dropout(x)
# add 2nd hidden layer, with relu activation function
x = self.fc2(x)
return x
#-#-# You so NOT have to modify the code below this line. #-#-#
# instantiate the CNN
model_scratch = Net()
# move tensors to GPU if CUDA is available
if use_cuda:
model_scratch.cuda()
In [70]:
model_scratch
Out[70]:
Net( (conv1): Conv2d(3, 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (conv2): Conv2d(16, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (conv3): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) (fc1): Linear(in_features=50176, out_features=500, bias=True) (fc2): Linear(in_features=500, out_features=133, bias=True) (dropout): Dropout(p=0.5) )
Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.
Answer: First i wanted CNN layers and so used Conv2d for setting three layers of conv layers namely (conv1): Conv2d(3, 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (conv2): Conv2d(16, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (conv3): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) since the images have 3 color channels, so in_channel of first conv layer namely conv1 will be "3" and then other channels and filter size are set by choosing arbitarily because they are hyperparameters and conv1 in_channel was data dependent parameter and so is last layer fc2 out_features as "133" due to 133 categories of dog breeds we want to classify.
Now, in_features of fc1 also is data dependent parameter and hence choosen as depth set by out_channels of conv3 and height * width set by pooling.
Pooling of window size 22 is used here to downsample the parameters otherwise training the network with much many parameters would require high computational complexity. So, after every convolutional layer and pooling , the dimensions height \ width is reduced by factor of 1/2. Max Pooling is used here so that maximum number out of all calculated numbers in feature maps is chosen so as to select relevant features only.
(pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) Dropout is used here to reduce overfitting as i saw the network tried to overfit the dataset very soon.
(dropout): Dropout(p=0.5, inplace=False)
(IMPLEMENTATION) Specify Loss Function and Optimizer¶
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.
In [73]:
import torch.optim as optim
### TODO: select loss function
criterion_scratch= nn.CrossEntropyLoss()
### TODO: select optimizer
optimizer_scratch = optim.SGD(model_scratch.parameters(), lr=0.002, momentum=0.9)
In [77]:
loaders_scratch = {
'train': trainloader,
'valid': validloader,
'test': testloader
}
(IMPLEMENTATION) Train and Validate the Model¶
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.
In [78]:
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
"""returns trained model"""
# initialize tracker for minimum validation loss
valid_loss_min = np.Inf
for epoch in range(1, n_epochs+1):
# initialize variables to monitor training and validation loss
train_loss = 0.0
valid_loss = 0.0
###################
# train the model #
###################
model.train()
for batch_idx, (data, target) in enumerate(loaders['train']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# clear the gradients of all optimized variables
optimizer.zero_grad()
## find the loss and update the model parameters accordingly
output = model(data)
loss = criterion(output, target)
loss.backward()
optimizer.step()
## record the average training loss, using something like
## train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
######################
# validate the model #
######################
model.eval()
for batch_idx, (data, target) in enumerate(loaders['valid']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
## update the average validation loss
#forward pass: compute predicted outputs by passing inputs to the model
output = model(data)
# calculate the batch loss
loss = criterion(output, target)
# update average validation loss
valid_loss = valid_loss + ((1 / (batch_idx + 1)) * (loss.data - valid_loss))
# print training/validation statistics
print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(epoch, train_loss,valid_loss))
## TODO: save the model if validation loss has decreased
if valid_loss <= valid_loss_min:
print('Validation loss decreased ({:.6f} --> {:.6f}). Saving model ...'.format(valid_loss_min,valid_loss))
print(" ")
torch.save(model.state_dict(),save_path)
valid_loss_min = valid_loss
# return trained model
return model
# train the model
model_scratch = train(50, loaders_scratch, model_scratch, optimizer_scratch,
criterion_scratch, use_cuda, 'model_scratch.pt')
# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
Epoch: 1 Training Loss: 4.886454 Validation Loss: 4.869884 Validation loss decreased (inf --> 4.869884). Saving model ... Epoch: 2 Training Loss: 4.870710 Validation Loss: 4.846887 Validation loss decreased (4.869884 --> 4.846887). Saving model ... Epoch: 3 Training Loss: 4.845342 Validation Loss: 4.790235 Validation loss decreased (4.846887 --> 4.790235). Saving model ... Epoch: 4 Training Loss: 4.774395 Validation Loss: 4.682907 Validation loss decreased (4.790235 --> 4.682907). Saving model ... Epoch: 5 Training Loss: 4.693630 Validation Loss: 4.579301 Validation loss decreased (4.682907 --> 4.579301). Saving model ... Epoch: 6 Training Loss: 4.631251 Validation Loss: 4.475163 Validation loss decreased (4.579301 --> 4.475163). Saving model ... Epoch: 7 Training Loss: 4.592149 Validation Loss: 4.390450 Validation loss decreased (4.475163 --> 4.390450). Saving model ... Epoch: 8 Training Loss: 4.559964 Validation Loss: 4.383087 Validation loss decreased (4.390450 --> 4.383087). Saving model ... Epoch: 9 Training Loss: 4.556668 Validation Loss: 4.354376 Validation loss decreased (4.383087 --> 4.354376). Saving model ... Epoch: 10 Training Loss: 4.515702 Validation Loss: 4.334084 Validation loss decreased (4.354376 --> 4.334084). Saving model ... Epoch: 11 Training Loss: 4.491201 Validation Loss: 4.307446 Validation loss decreased (4.334084 --> 4.307446). Saving model ... Epoch: 12 Training Loss: 4.494987 Validation Loss: 4.294216 Validation loss decreased (4.307446 --> 4.294216). Saving model ... Epoch: 13 Training Loss: 4.458324 Validation Loss: 4.251158 Validation loss decreased (4.294216 --> 4.251158). Saving model ... Epoch: 14 Training Loss: 4.431738 Validation Loss: 4.278013 Epoch: 15 Training Loss: 4.431150 Validation Loss: 4.322583 Epoch: 16 Training Loss: 4.397852 Validation Loss: 4.222128 Validation loss decreased (4.251158 --> 4.222128). Saving model ... Epoch: 17 Training Loss: 4.375138 Validation Loss: 4.265685 Epoch: 18 Training Loss: 4.374236 Validation Loss: 4.199364 Validation loss decreased (4.222128 --> 4.199364). Saving model ... Epoch: 19 Training Loss: 4.348927 Validation Loss: 4.135077 Validation loss decreased (4.199364 --> 4.135077). Saving model ... Epoch: 20 Training Loss: 4.327661 Validation Loss: 4.178394 Epoch: 21 Training Loss: 4.317808 Validation Loss: 4.204782 Epoch: 22 Training Loss: 4.284447 Validation Loss: 4.237369 Epoch: 23 Training Loss: 4.282736 Validation Loss: 4.065900 Validation loss decreased (4.135077 --> 4.065900). Saving model ... Epoch: 24 Training Loss: 4.246348 Validation Loss: 4.141339 Epoch: 25 Training Loss: 4.248358 Validation Loss: 4.161367 Epoch: 26 Training Loss: 4.219202 Validation Loss: 4.118932 Epoch: 27 Training Loss: 4.217691 Validation Loss: 4.042978 Validation loss decreased (4.065900 --> 4.042978). Saving model ... Epoch: 28 Training Loss: 4.188429 Validation Loss: 4.053004 Epoch: 29 Training Loss: 4.197796 Validation Loss: 4.067964 Epoch: 30 Training Loss: 4.171879 Validation Loss: 4.014311 Validation loss decreased (4.042978 --> 4.014311). Saving model ... Epoch: 31 Training Loss: 4.129994 Validation Loss: 3.964827 Validation loss decreased (4.014311 --> 3.964827). Saving model ... Epoch: 32 Training Loss: 4.113400 Validation Loss: 4.065144 Epoch: 33 Training Loss: 4.100185 Validation Loss: 3.918233 Validation loss decreased (3.964827 --> 3.918233). Saving model ... Epoch: 34 Training Loss: 4.112092 Validation Loss: 3.951040 Epoch: 35 Training Loss: 4.046886 Validation Loss: 3.923172 Epoch: 36 Training Loss: 4.040989 Validation Loss: 3.946170 Epoch: 37 Training Loss: 4.029156 Validation Loss: 3.842577 Validation loss decreased (3.918233 --> 3.842577). Saving model ... Epoch: 38 Training Loss: 4.000926 Validation Loss: 3.826622 Validation loss decreased (3.842577 --> 3.826622). Saving model ... Epoch: 39 Training Loss: 3.992855 Validation Loss: 3.817704 Validation loss decreased (3.826622 --> 3.817704). Saving model ... Epoch: 40 Training Loss: 3.974399 Validation Loss: 3.856467 Epoch: 41 Training Loss: 3.951888 Validation Loss: 3.787260 Validation loss decreased (3.817704 --> 3.787260). Saving model ... Epoch: 42 Training Loss: 3.933193 Validation Loss: 3.790864 Epoch: 43 Training Loss: 3.934397 Validation Loss: 3.754840 Validation loss decreased (3.787260 --> 3.754840). Saving model ... Epoch: 44 Training Loss: 3.923846 Validation Loss: 3.754156 Validation loss decreased (3.754840 --> 3.754156). Saving model ... Epoch: 45 Training Loss: 3.882241 Validation Loss: 3.704870 Validation loss decreased (3.754156 --> 3.704870). Saving model ... Epoch: 46 Training Loss: 3.855513 Validation Loss: 3.791137 Epoch: 47 Training Loss: 3.832333 Validation Loss: 3.703926 Validation loss decreased (3.704870 --> 3.703926). Saving model ... Epoch: 48 Training Loss: 3.854392 Validation Loss: 3.641114 Validation loss decreased (3.703926 --> 3.641114). Saving model ... Epoch: 49 Training Loss: 3.813023 Validation Loss: 3.609769 Validation loss decreased (3.641114 --> 3.609769). Saving model ... Epoch: 50 Training Loss: 3.844024 Validation Loss: 3.760702
In [111]:
x=torch.load('model_scratch.pt')
In [112]:
print(x)
OrderedDict([('conv1.weight', tensor([[[[ 0.0465, 0.1565, 0.2504],
[-0.1840, -0.0939, 0.1578],
[-0.0131, -0.0901, 0.1793]],
[[-0.2324, -0.2807, -0.1252],
[-0.2520, -0.2213, -0.0824],
[-0.2972, 0.0586, 0.1809]],
[[ 0.1706, -0.0393, -0.0340],
[-0.0116, 0.0891, 0.2759],
[ 0.0045, 0.2198, -0.0445]]],
[[[-0.1015, 0.1372, 0.0787],
[-0.2375, 0.1223, 0.2429],
[-0.1787, -0.1256, 0.2642]],
[[ 0.0174, -0.1292, -0.1294],
[ 0.0955, 0.0566, 0.2900],
[ 0.0623, 0.1976, 0.3298]],
[[ 0.0118, -0.2195, -0.0498],
[-0.1853, 0.0131, -0.1021],
[-0.2037, -0.0328, 0.0041]]],
[[[-0.1741, 0.0299, 0.1170],
[-0.0475, -0.1132, 0.0349],
[ 0.0677, -0.0034, 0.0705]],
[[-0.0355, 0.2258, 0.1169],
[-0.0813, 0.1380, 0.0910],
[ 0.0340, 0.3123, 0.1738]],
[[-0.2314, -0.2223, -0.0911],
[-0.1742, -0.1220, -0.1017],
[ 0.0881, 0.0790, 0.1748]]],
[[[ 0.2754, -0.0807, 0.0625],
[ 0.1614, 0.2688, -0.0596],
[ 0.0013, 0.0521, 0.2176]],
[[-0.0884, -0.0486, 0.1859],
[-0.0887, 0.0912, 0.0917],
[-0.1149, 0.1913, 0.1664]],
[[ 0.1040, -0.2980, -0.1277],
[-0.2182, 0.0659, -0.3152],
[-0.0684, -0.0971, -0.3381]]],
[[[ 0.1703, 0.1951, 0.2575],
[ 0.3504, 0.2571, 0.3384],
[ 0.2744, 0.2287, 0.0296]],
[[-0.2272, 0.0375, -0.1801],
[-0.2633, -0.1294, -0.2626],
[-0.0349, -0.1868, 0.0389]],
[[ 0.0703, -0.2359, -0.1021],
[-0.1599, -0.2192, 0.0603],
[-0.2762, -0.0543, -0.0945]]],
[[[ 0.0546, -0.0891, 0.1242],
[ 0.0591, -0.1415, 0.1013],
[ 0.1711, 0.0966, -0.0785]],
[[-0.0940, 0.1662, -0.1753],
[ 0.0744, -0.1899, 0.1223],
[ 0.0684, -0.0574, 0.1063]],
[[-0.1044, 0.1448, -0.1585],
[-0.0618, -0.0621, 0.1345],
[ 0.0311, 0.0711, 0.0967]]],
[[[-0.1304, 0.0621, 0.3157],
[-0.1381, -0.0203, 0.0972],
[-0.2772, -0.1845, 0.1211]],
[[-0.2004, -0.1397, 0.3037],
[ 0.1113, -0.1062, 0.2262],
[-0.2267, -0.1320, 0.2492]],
[[ 0.0313, 0.1834, 0.2418],
[-0.1259, -0.1976, 0.0223],
[-0.2057, -0.0073, 0.1967]]],
[[[ 0.3026, -0.1045, 0.1638],
[-0.0263, -0.2005, 0.3074],
[-0.1548, -0.0558, -0.1889]],
[[-0.1034, -0.1183, -0.1607],
[-0.1480, -0.2983, 0.1223],
[-0.0265, -0.3556, 0.0318]],
[[ 0.1163, 0.1647, 0.1812],
[ 0.2081, -0.0723, 0.1313],
[ 0.2270, -0.1294, -0.0062]]],
[[[-0.0695, 0.2137, -0.0411],
[ 0.1077, -0.0160, 0.0742],
[ 0.1167, 0.1976, -0.0936]],
[[-0.1786, 0.1585, -0.2041],
[-0.0130, -0.2031, -0.0758],
[-0.1807, -0.0678, 0.1135]],
[[-0.2093, -0.2736, -0.0632],
[ 0.0055, 0.0009, -0.1975],
[-0.1364, -0.2588, -0.1082]]],
[[[-0.0326, -0.0445, 0.1354],
[-0.1757, -0.1146, -0.0000],
[-0.1532, 0.1123, -0.1134]],
[[ 0.2630, 0.2555, 0.0020],
[ 0.2487, 0.2594, 0.1679],
[ 0.2379, 0.0218, 0.1263]],
[[-0.1572, -0.1472, -0.0209],
[ 0.0150, 0.1018, -0.1596],
[ 0.1252, 0.1232, 0.1218]]],
[[[ 0.2321, -0.1713, -0.0722],
[ 0.2527, -0.2040, 0.0058],
[ 0.1666, 0.1085, -0.2209]],
[[ 0.1508, 0.1943, -0.1928],
[ 0.2012, -0.2271, -0.2502],
[ 0.2460, 0.1619, -0.2219]],
[[ 0.2358, 0.0519, -0.0689],
[ 0.3014, -0.2270, -0.2060],
[ 0.1653, -0.0843, -0.2639]]],
[[[ 0.0478, -0.0525, -0.1984],
[-0.2181, -0.2584, -0.0418],
[ 0.2235, -0.0735, 0.1713]],
[[ 0.3212, 0.1353, 0.2163],
[ 0.0281, 0.2578, 0.3492],
[ 0.4378, 0.3022, 0.0776]],
[[-0.0617, -0.2045, -0.2323],
[-0.0468, -0.3203, -0.3295],
[-0.0985, -0.2355, -0.1230]]],
[[[-0.0777, 0.1794, 0.1692],
[ 0.0848, 0.0677, -0.0954],
[ 0.0910, 0.0362, 0.0100]],
[[-0.1310, -0.0188, -0.0918],
[-0.1489, -0.1076, -0.2229],
[-0.0605, -0.1542, 0.1301]],
[[ 0.1197, 0.1407, 0.1286],
[ 0.2305, 0.1590, 0.2054],
[ 0.1429, 0.1599, 0.0477]]],
[[[ 0.2837, 0.2729, -0.0177],
[ 0.1154, 0.3126, 0.1698],
[ 0.2452, 0.1659, 0.1603]],
[[-0.0856, -0.2465, -0.0966],
[-0.2664, -0.1515, -0.0705],
[-0.2093, -0.1784, -0.0731]],
[[ 0.0014, 0.0024, -0.1467],
[ 0.0993, -0.1709, -0.0846],
[ 0.0062, -0.1315, 0.0047]]],
[[[ 0.1173, 0.1662, 0.1431],
[ 0.1836, 0.1215, 0.1038],
[ 0.2411, 0.2038, 0.0494]],
[[-0.1672, -0.2300, -0.0635],
[-0.0196, -0.1792, -0.2139],
[-0.2453, 0.0353, -0.0084]],
[[-0.1196, 0.1167, -0.0559],
[-0.0387, 0.0842, 0.2007],
[-0.2028, -0.0835, -0.1024]]],
[[[ 0.0582, -0.0218, 0.1601],
[ 0.0828, -0.1436, 0.1276],
[ 0.1245, -0.1438, 0.0877]],
[[-0.1204, -0.2251, -0.1835],
[-0.2120, -0.1200, -0.0816],
[-0.1773, -0.1447, -0.0688]],
[[ 0.0210, 0.0102, 0.2308],
[-0.0542, -0.0499, 0.2065],
[-0.1309, 0.2067, 0.0775]]]], device='cuda:0')), ('conv1.bias', tensor([-0.0993, -0.1771, -0.1509, 0.4545, 0.0958, 0.3514, 0.2067,
0.1312, -0.1388, -0.1153, -0.0170, 0.2880, 0.2046, -0.0469,
-0.0879, 0.1169], device='cuda:0')), ('conv2.weight', tensor([[[[-0.0264, -0.0031, -0.0594],
[-0.0043, 0.0354, -0.0365],
[ 0.0417, -0.0585, -0.0620]],
[[-0.0167, 0.0812, 0.0015],
[ 0.0658, -0.0671, -0.0427],
[ 0.0664, -0.0129, 0.0031]],
[[-0.0158, -0.0325, 0.0492],
[-0.0138, -0.0161, -0.0400],
[ 0.0353, 0.0138, -0.0557]],
...,
[[-0.0445, -0.0069, 0.0952],
[ 0.0080, 0.0407, 0.0396],
[ 0.0802, 0.0509, 0.0166]],
[[ 0.0485, 0.0932, 0.0211],
[-0.0061, -0.0127, 0.0084],
[ 0.0920, 0.0241, -0.0202]],
[[-0.1097, -0.0618, 0.0605],
[-0.0433, -0.0508, -0.0274],
[-0.0175, -0.0124, -0.0933]]],
[[[ 0.0427, -0.0621, -0.0320],
[-0.0257, -0.0738, -0.0613],
[ 0.0765, 0.0159, 0.0732]],
[[ 0.0010, 0.0257, -0.0497],
[ 0.0696, -0.0583, -0.0129],
[-0.0486, 0.0166, 0.0532]],
[[-0.0490, 0.0631, -0.0656],
[-0.0297, 0.0542, -0.0235],
[ 0.0550, 0.0726, -0.0415]],
...,
[[ 0.0442, 0.0267, 0.0278],
[ 0.0014, -0.1112, -0.0286],
[-0.0193, -0.0753, -0.0383]],
[[-0.0692, 0.0054, 0.0533],
[-0.0827, -0.0062, -0.0802],
[-0.1001, -0.0623, -0.0931]],
[[-0.0354, 0.0537, 0.1008],
[ 0.0363, -0.0158, 0.0887],
[ 0.0342, 0.0782, 0.0599]]],
[[[-0.0606, -0.0978, 0.0604],
[-0.0690, 0.0728, -0.0180],
[-0.0416, 0.0804, 0.0643]],
[[-0.0101, -0.0377, 0.0050],
[ 0.0335, -0.0545, 0.0197],
[ 0.0051, 0.0878, -0.0695]],
[[ 0.0482, -0.0524, 0.0680],
[-0.0794, -0.0735, -0.0853],
[ 0.0566, 0.0554, 0.0828]],
...,
[[-0.0252, 0.0245, -0.0242],
[-0.0429, 0.0069, -0.0970],
[ 0.0067, -0.0123, -0.0775]],
[[ 0.0074, -0.0888, -0.0851],
[-0.0609, -0.1148, 0.0098],
[-0.0261, -0.0202, 0.0181]],
[[-0.0467, 0.0355, -0.0301],
[-0.0842, -0.0834, -0.0814],
[ 0.0574, -0.0212, -0.0956]]],
...,
[[[ 0.0118, 0.0940, 0.0110],
[ 0.0514, 0.0958, 0.1071],
[ 0.0089, 0.0872, -0.0310]],
[[ 0.0169, -0.0675, -0.0143],
[-0.0056, 0.0015, 0.0467],
[ 0.0235, -0.0162, 0.0162]],
[[-0.0930, 0.0480, -0.0054],
[-0.0802, -0.0285, 0.0841],
[ 0.0093, -0.0534, -0.0158]],
...,
[[ 0.1275, 0.0455, 0.1294],
[ 0.0643, 0.1173, 0.0138],
[ 0.0923, -0.0057, 0.1068]],
[[ 0.0972, -0.0159, 0.0838],
[ 0.1279, 0.0209, 0.1340],
[ 0.1098, 0.0115, 0.0472]],
[[ 0.0498, 0.0820, 0.0794],
[ 0.0266, 0.0693, 0.0888],
[ 0.0075, 0.0665, 0.0590]]],
[[[ 0.0020, -0.0575, -0.0218],
[ 0.0349, -0.1086, -0.0208],
[-0.0292, 0.0299, -0.1087]],
[[-0.0013, 0.0677, 0.0565],
[-0.0301, 0.1143, 0.0654],
[-0.0544, 0.0118, 0.0220]],
[[-0.0150, 0.0114, 0.0797],
[ 0.0726, -0.0278, 0.0170],
[-0.0380, -0.0427, -0.0570]],
...,
[[ 0.0800, 0.0711, 0.0772],
[-0.0324, 0.0414, -0.0431],
[ 0.0948, 0.1036, -0.0316]],
[[-0.0141, -0.0306, 0.0457],
[ 0.0306, 0.0867, 0.0419],
[-0.0530, 0.0325, 0.0681]],
[[-0.0807, -0.1252, -0.0462],
[-0.0489, -0.0378, -0.0873],
[ 0.0384, -0.0984, -0.0490]]],
[[[ 0.0009, -0.0090, -0.0482],
[ 0.0442, -0.0993, -0.0112],
[ 0.0418, -0.0291, -0.0688]],
[[ 0.0104, -0.0165, -0.0140],
[-0.0299, 0.0200, -0.0208],
[-0.0155, -0.0321, -0.0449]],
[[-0.0758, 0.0354, 0.0021],
[-0.0531, -0.0873, -0.0189],
[-0.0583, -0.0600, -0.0177]],
...,
[[ 0.0776, 0.0425, -0.0546],
[-0.0064, 0.0111, -0.0614],
[-0.0433, 0.0423, 0.0369]],
[[ 0.0545, -0.0806, -0.0109],
[-0.0110, 0.0175, -0.0741],
[-0.0330, -0.0868, 0.0650]],
[[ 0.0301, 0.0300, 0.0856],
[-0.0161, -0.0405, 0.0077],
[ 0.0499, -0.0590, 0.1165]]]], device='cuda:0')), ('conv2.bias', tensor([ 0.0320, 0.0326, 0.1237, 0.0532, 0.1941, -0.0275, -0.0434,
-0.0525, -0.1105, 0.0970, -0.2421, -0.2137, 0.1203, 0.1710,
0.0058, 0.1122, 0.3066, -0.0443, 0.1097, 0.1720, 0.1554,
-0.2849, -0.0270, 0.0520, -0.0545, 0.1140, 0.0999, 0.0807,
0.2076, -0.2372, -0.0824, 0.0867], device='cuda:0')), ('conv3.weight', tensor([[[[ 4.6852e-02, 5.0486e-02, 4.3937e-02],
[-5.7202e-03, -3.3431e-02, -1.4087e-02],
[-3.8967e-02, 4.4420e-02, 2.0374e-02]],
[[-5.1399e-02, -1.7822e-02, 3.7564e-02],
[-5.1139e-02, 4.0879e-02, -2.7343e-03],
[ 2.8991e-03, -8.7243e-03, 2.0934e-03]],
[[ 2.8301e-02, -1.1863e-02, 2.9968e-02],
[-1.1567e-02, -1.0058e-02, 3.0667e-02],
[ 4.0002e-02, 2.5887e-03, -7.0584e-02]],
...,
[[-2.2318e-02, 5.3233e-03, -2.3300e-02],
[ 4.3330e-02, -2.0154e-02, -2.3378e-02],
[-2.6462e-02, -5.4898e-02, 4.3116e-02]],
[[-6.2134e-03, 1.3455e-02, -2.6765e-02],
[-1.3288e-02, -2.4298e-03, 1.2675e-02],
[-5.4932e-02, -3.0544e-03, 4.0381e-02]],
[[ 2.7495e-02, 1.2049e-02, -3.5430e-02],
[-4.8138e-02, 1.9761e-02, -7.5945e-02],
[-3.5500e-02, -5.4426e-02, -3.7988e-02]]],
[[[ 2.9985e-02, -2.9758e-02, 5.4989e-02],
[-4.7337e-02, -2.3797e-05, -2.1829e-02],
[ 7.1557e-03, 5.1508e-02, 1.8769e-02]],
[[-3.3178e-02, 2.9178e-02, 1.1794e-02],
[ 4.0939e-02, -5.0881e-02, 2.1792e-02],
[ 5.5659e-02, -1.7892e-02, -1.9852e-02]],
[[ 4.2256e-02, 6.2801e-03, 4.0052e-02],
[ 9.2755e-03, 2.0323e-02, -3.5524e-02],
[ 2.8544e-02, -2.6119e-02, -1.5573e-03]],
...,
[[-3.9466e-02, 4.7164e-02, 3.2267e-02],
[-2.3447e-02, -1.7347e-02, -3.1937e-02],
[ 1.9844e-02, -2.1117e-02, 5.7599e-03]],
[[ 1.4794e-02, 2.0416e-02, 3.4824e-02],
[ 2.4939e-02, 7.1409e-02, 2.7323e-02],
[ 2.5756e-03, -1.5360e-02, 4.0678e-02]],
[[ 2.4739e-03, -7.0198e-03, -8.0871e-02],
[ 1.3554e-02, -1.7697e-02, -3.8826e-02],
[-3.1188e-02, 7.4285e-03, -4.0658e-02]]],
[[[ 1.6213e-02, -3.9243e-02, 3.0313e-03],
[-7.8191e-02, -4.3383e-02, -2.2395e-02],
[ 1.6752e-02, -4.1560e-02, -6.8602e-02]],
[[-6.4614e-02, 4.6610e-02, -6.1750e-02],
[ 4.9193e-02, -6.2214e-03, 4.5159e-02],
[-2.8385e-02, 2.2691e-02, 1.2832e-02]],
[[ 3.8315e-02, -3.2556e-02, -3.4624e-02],
[-2.5565e-02, 3.6747e-02, 3.0933e-02],
[-1.4258e-02, -2.2455e-02, -6.0957e-02]],
...,
[[ 4.3711e-02, 2.8191e-02, 3.8118e-02],
[ 4.3376e-02, -3.3189e-02, 3.1494e-02],
[ 3.2973e-02, -5.1448e-02, -4.9436e-02]],
[[ 2.1458e-02, -1.0731e-02, -4.9346e-02],
[ 3.0597e-03, 1.2179e-02, -1.7936e-02],
[-1.5766e-02, -8.7709e-03, -4.3982e-02]],
[[-4.7058e-02, -2.8989e-02, 7.3591e-03],
[ 9.9980e-03, -3.3280e-03, -8.9537e-03],
[-6.9904e-03, 1.8197e-02, 5.0580e-02]]],
...,
[[[-9.2505e-03, -2.3149e-02, 3.1951e-02],
[ 1.5903e-02, -6.0548e-02, -2.9010e-02],
[-3.7738e-02, -3.0553e-02, 3.3435e-02]],
[[ 4.6925e-02, -6.6744e-02, 3.8296e-02],
[-4.1004e-02, 4.6755e-02, -5.3512e-02],
[-7.4257e-03, 4.1647e-02, -3.2204e-02]],
[[-4.1615e-02, 1.9807e-02, 6.3452e-02],
[ 2.6531e-02, 3.8398e-02, 1.0671e-02],
[-3.1353e-02, 5.5716e-02, 4.6405e-02]],
...,
[[-1.8952e-02, 5.3693e-03, -2.6920e-02],
[ 1.2797e-02, -4.8275e-02, 1.2308e-02],
[ 2.7460e-02, -3.3120e-02, -4.3686e-02]],
[[-3.2667e-02, 4.5487e-02, -4.3000e-02],
[ 1.8871e-02, 1.3514e-02, -2.5856e-02],
[ 4.2529e-02, 3.8219e-02, -2.7862e-02]],
[[ 1.6359e-02, -2.3742e-02, 2.2513e-02],
[-1.8460e-05, -2.7297e-02, 4.8577e-02],
[ 3.4348e-02, 4.6764e-02, -2.3108e-02]]],
[[[-2.3361e-02, 1.5465e-02, 4.0832e-02],
[-2.9136e-02, 2.9451e-02, 4.7394e-02],
[ 1.7724e-02, 1.5922e-03, -7.7373e-03]],
[[ 1.9995e-02, -4.4331e-02, 5.6403e-02],
[ 2.7596e-02, 3.5516e-02, -1.7756e-02],
[ 4.7051e-02, 3.4569e-02, -1.1666e-02]],
[[-1.2025e-03, -5.5049e-02, 2.2141e-02],
[-5.4600e-02, -2.1611e-02, -3.9448e-02],
[-4.3274e-02, 1.1227e-02, 1.3548e-02]],
...,
[[ 4.2225e-02, 6.2906e-02, -7.2989e-03],
[ 1.0809e-02, -6.2649e-03, -2.5212e-03],
[-1.2451e-02, -2.6574e-02, -3.3746e-02]],
[[-3.2300e-02, 3.3959e-02, -1.9392e-02],
[ 4.8953e-02, -5.1886e-02, -5.7255e-02],
[ 2.4199e-02, -2.1482e-02, 4.0496e-02]],
[[-7.5513e-03, 4.9607e-02, -1.8083e-03],
[-1.6868e-02, 2.8467e-03, -1.4762e-02],
[ 2.1030e-02, -1.6926e-02, -5.1921e-02]]],
[[[ 1.4528e-02, -7.0059e-03, 1.4013e-02],
[ 1.8705e-02, -1.3924e-02, 1.7836e-02],
[-2.2886e-02, 5.2258e-03, -3.2875e-02]],
[[-3.5846e-02, -4.5228e-02, -5.6458e-02],
[-5.1115e-02, 6.0128e-03, 2.4131e-02],
[-3.4686e-02, -5.7048e-03, -4.7735e-02]],
[[-2.3231e-02, -6.6645e-02, -3.4148e-02],
[-5.3877e-03, -6.3311e-02, -7.9108e-02],
[ 2.1130e-02, -2.7885e-03, 1.6616e-02]],
...,
[[-6.8189e-02, -3.0251e-02, -1.1770e-02],
[ 1.5548e-02, 2.1651e-02, -5.8069e-02],
[-6.6048e-02, -1.1156e-02, -4.8165e-02]],
[[-1.8164e-02, -7.4838e-02, -7.3241e-02],
[ 3.7362e-02, -2.7637e-02, 9.1749e-03],
[-1.1670e-02, -7.0548e-02, -3.8337e-02]],
[[ 6.3369e-02, 6.5511e-02, -1.8001e-02],
[ 6.1500e-02, -2.1467e-02, 4.8199e-02],
[ 8.2496e-02, 3.7272e-02, 3.2286e-03]]]], device='cuda:0')), ('conv3.bias', tensor([ 0.0784, -0.0973, -0.0061, 0.1303, -0.0297, 0.0261, 0.0479,
0.0834, -0.0841, 0.0405, 0.1520, -0.0896, 0.1331, 0.0771,
0.0449, -0.0259, 0.0270, -0.0030, 0.1155, 0.1150, -0.0631,
0.1213, 0.1363, -0.2460, -0.0429, -0.0481, 0.1316, -0.0386,
-0.1073, -0.0461, -0.0811, 0.0904, -0.0108, -0.0381, -0.0646,
-0.0770, 0.0438, 0.0062, -0.0487, 0.0059, -0.0991, -0.0162,
0.0879, -0.0619, 0.0661, 0.0942, -0.0654, -0.0538, -0.0604,
-0.0595, 0.0782, -0.0556, -0.1349, -0.1006, -0.0822, -0.0512,
0.2315, 0.0059, 0.1961, -0.0119, 0.1239, -0.0193, -0.0364,
0.0990], device='cuda:0')), ('fc1.weight', tensor([[-3.6689e-03, -3.6102e-03, 6.7652e-04, ..., -1.1955e-03,
4.0622e-03, 4.1334e-03],
[ 8.8914e-03, -2.2049e-03, 1.5926e-03, ..., -2.0150e-03,
-3.8095e-03, -1.9596e-03],
[ 2.4739e-03, -3.2087e-03, 8.4674e-04, ..., 2.2668e-03,
-3.2802e-03, 5.8588e-04],
...,
[ 4.2320e-03, -2.6308e-03, 7.1230e-04, ..., 1.0020e-03,
1.8456e-03, -8.3567e-05],
[-1.6615e-03, -7.7622e-05, 3.5936e-03, ..., 3.7282e-03,
2.3514e-04, -2.8572e-03],
[ 2.4475e-03, 8.9700e-04, -1.3541e-03, ..., 5.4399e-03,
4.3276e-03, 5.3196e-04]], device='cuda:0')), ('fc1.bias', tensor(1.00000e-02 *
[-0.1290, 0.0972, 0.3900, 0.4664, -0.0366, 0.2806, -0.3355,
0.1853, 0.6141, 0.1089, -0.4953, -0.4456, -0.0249, -0.2706,
-0.3820, -0.4339, -0.3704, -0.1601, -0.0625, -0.4860, 0.1661,
0.0051, -0.6044, -0.3081, -0.0912, -0.1486, -0.9694, 0.4826,
0.2760, -0.3705, -0.3978, 0.0651, -0.5815, 0.0072, 0.2880,
-0.0049, -0.7420, 0.2073, 0.5366, 0.3279, 0.5251, 0.8175,
0.7506, 0.4857, 1.0641, -1.0098, -0.3269, -0.5646, 0.2917,
0.1036, 0.8856, -0.2103, 0.2956, 0.1940, 0.1229, 0.2626,
0.0348, -0.2476, 0.3644, 0.4362, -0.5192, 0.2103, 0.7171,
0.1623, 0.3109, -0.4751, 0.1073, 0.4012, -0.2645, 0.9132,
0.2057, 0.5919, -0.8151, -0.0685, 0.0352, -0.4837, -0.2418,
0.2706, 0.2759, -0.5072, 0.5504, -0.3575, -0.1709, -0.0306,
-0.4082, 0.4621, 0.2519, -0.1903, 0.0982, -0.2338, 0.4151,
0.4511, 0.0979, 0.1871, 0.2887, -0.5288, -0.2455, -0.7655,
-0.1072, 0.3098, 0.1118, 0.4183, 0.0252, 0.0923, 0.2922,
0.5617, -1.2385, 0.4288, 0.0151, -1.0380, -0.0637, -0.5908,
0.2663, 0.1269, 0.2043, -0.0431, -0.1429, -0.0792, 0.1501,
-0.0721, 0.3271, 0.0790, 0.3055, 0.2835, 0.0862, 0.0375,
-0.8794, 0.1481, 0.4849, -0.2210, 0.3049, 0.1685, -0.4693,
-0.2945, -0.1142, -0.7642, 0.2400, -0.3950, 0.2606, 0.7142,
-0.8298, -1.2596, -0.3443, 0.2966, 0.4811, -0.0549, 0.2240,
-0.4178, -1.0925, 0.4907, 0.2264, 0.9542, 0.0442, 0.3403,
0.4242, 0.0905, 0.3551, 1.1507, -0.1288, 0.5225, -0.2750,
-0.4774, 0.6015, 0.3932, -0.0547, 0.8840, 0.7113, -0.0154,
-0.0078, -0.8654, 0.8061, -0.0133, 0.0981, -0.5147, -0.3757,
0.6186, 1.0828, -0.1341, -0.0082, -0.4142, 0.0377, -0.0646,
-0.3954, 0.6165, -1.3002, 0.6185, 0.2116, -0.4660, 0.4681,
0.3015, -0.3506, -0.4434, 0.0727, 0.3395, 0.0889, 0.9308,
-0.3832, -0.0393, 0.5366, 0.3674, -0.1935, 0.0884, -0.2851,
0.2170, -0.2964, -0.0130, 0.1921, 0.0419, -0.5431, 0.1323,
0.7462, 0.2788, -0.0636, 0.1865, 0.0942, 0.2729, 0.1787,
-0.3952, 0.4281, -0.3310, -0.4528, 0.1729, -0.1254, 0.1198,
-0.2045, 0.2259, 0.7042, -0.0524, 0.3022, -0.1410, -0.6348,
-0.4411, -0.2735, -0.2171, 0.8296, -0.3628, 1.6227, 0.4329,
-0.5519, 0.9201, 0.7639, 0.8409, 0.3761, 0.3687, -0.1616,
0.2991, -0.0816, -0.5134, 0.1363, -0.0217, 0.7903, 0.4045,
-0.2387, 0.0100, 0.6653, 0.5910, 0.6267, -0.1937, -0.3233,
0.1929, -0.5042, -0.1491, 0.2803, -0.1864, 0.1886, -0.3481,
0.2452, 0.0012, -1.0487, -0.1028, -0.1403, 1.0825, -0.7912,
0.1980, -0.1379, 0.2452, -0.8398, 0.0698, -0.5123, -0.5717,
-0.4590, 0.7903, -0.2579, 0.3082, -0.2167, -0.3014, 0.5299,
0.2972, 0.4104, 0.1482, 0.1237, -0.1217, 1.0255, -0.4438,
0.8488, -0.1845, -0.0388, -0.2549, -0.5512, -0.0175, -0.5637,
-0.0603, 0.5113, 0.3510, -0.1762, 0.0383, 0.2118, 0.3587,
-0.2088, -0.0263, 0.1839, -0.4262, -0.4049, -1.3449, -0.2664,
-0.4917, -0.3426, 0.2489, 0.9667, -0.2633, -0.3552, -0.0860,
-0.0669, -0.4847, 0.0198, 0.2065, -0.2060, -0.4918, 0.2062,
0.2760, -0.0762, 0.1108, -0.0783, 0.9160, -0.1312, 0.6869,
-0.4527, -0.3586, -0.6067, -0.2937, 0.5736, 0.0451, -0.6966,
-0.1160, 0.1010, 0.2014, -0.1515, -0.1778, -0.1443, 0.4137,
-0.5445, -0.2527, 0.6189, 0.1347, -0.3689, -0.4325, 1.4922,
-0.1166, 0.2900, 0.1749, 0.1976, 0.3970, -0.6999, -0.9394,
0.0007, 0.0636, -0.4859, -0.4967, 0.2474, -0.2886, -0.0216,
0.2420, 0.1337, 0.1317, -0.4044, -0.3510, 0.0580, 0.3430,
-0.0944, 1.1259, -0.4642, 0.1483, -0.0119, -0.5231, 0.7040,
-0.6413, 0.2908, 0.7542, 0.0208, -0.3164, 0.3320, 0.6684,
-0.0505, 0.3689, -0.5876, -0.3874, -0.4348, -0.6184, 0.3327,
-0.5236, 0.1116, -0.5026, -0.1160, -0.5498, 0.0153, -0.1350,
-0.2876, -0.0477, 0.4372, 0.5754, 0.3288, -0.2666, 0.0183,
0.6479, 0.4525, 0.1838, -0.4237, -0.4134, -0.1334, -0.3451,
0.5127, 0.2302, 0.3508, 0.1338, -0.3829, -0.0535, -0.5875,
-0.2179, 0.2527, 0.7483, 0.3733, 0.0596, -0.1890, -0.2238,
0.1256, -0.0690, -0.1548, -0.6493, 0.0383, -0.4981, -0.3119,
-0.1673, -0.0500, -0.4920, 0.0722, 0.9436, 0.5837, -0.3610,
-0.2050, -0.2717, 0.7572, -0.0773, 0.1395, -0.0294, 0.8707,
-0.3874, -0.4790, 0.3776, -0.4535, 0.1233, 0.1259, 0.1719,
-0.8480, 0.7353, -0.4770, 0.0087, 0.3133, -0.3735, 0.0061,
0.1733, -0.4738, -0.6675, -0.7165, -0.0952, 0.1796, -0.3712,
-0.2079, 0.2426, 0.0148, 0.6824, 0.2144, 0.6646, -0.4311,
-0.1254, -0.5136, -0.3020, 0.3229, -0.3474, 1.0491, -0.1206,
0.1157, 0.1100, 0.8283, 0.2275, 0.3043, -0.0830, 0.1000,
0.0373, 0.2060, 0.6539], device='cuda:0')), ('fc2.weight', tensor([[-1.8818e-02, 7.4549e-02, 5.4544e-04, ..., -2.9327e-02,
-5.1068e-02, -5.1691e-02],
[-7.0447e-03, 2.4585e-02, -2.7321e-02, ..., 4.6821e-02,
-2.8857e-02, 4.4209e-02],
[-2.4085e-03, -5.9828e-02, 2.4129e-02, ..., 6.3012e-02,
-5.3033e-03, 4.1330e-02],
...,
[ 2.2512e-02, 5.8869e-02, -1.5026e-02, ..., -1.2647e-02,
2.5682e-02, 2.0514e-02],
[ 2.5133e-02, 1.5763e-02, 1.0079e-02, ..., -5.3694e-03,
1.0062e-02, 5.1652e-02],
[ 2.1872e-02, -9.2089e-03, 3.0830e-02, ..., 3.7944e-02,
-2.8961e-02, 2.4964e-03]], device='cuda:0')), ('fc2.bias', tensor([-0.0736, 0.0435, -0.0290, 0.0371, 0.0522, -0.0074, 0.0184,
-0.0127, 0.0208, 0.0441, 0.0549, 0.0761, 0.0039, 0.0444,
0.0130, 0.0024, 0.0023, 0.0233, 0.0305, 0.1126, 0.0367,
0.0078, -0.0074, -0.0157, -0.0406, -0.0764, 0.0160, 0.0042,
0.0185, 0.0446, 0.0273, 0.0067, -0.0136, 0.0010, 0.0247,
0.0620, 0.0268, 0.0436, 0.0326, 0.0467, 0.0380, 0.0434,
-0.0334, -0.0166, 0.0504, -0.0073, 0.0489, -0.0273, -0.0247,
0.0534, -0.0022, 0.0073, -0.0403, 0.0141, -0.0250, 0.0362,
-0.0856, -0.0034, -0.0110, 0.0162, -0.0611, 0.0301, -0.0326,
0.0296, -0.0252, -0.0333, -0.0489, -0.0233, 0.0212, 0.0015,
0.0556, 0.0225, 0.0017, -0.0028, -0.0030, 0.0074, -0.0288,
-0.0436, -0.0053, 0.0113, -0.0091, 0.0269, 0.0408, 0.0003,
0.0253, -0.0379, 0.0211, -0.0129, 0.0567, -0.0073, -0.0495,
0.0079, -0.0115, -0.0484, 0.0113, -0.0399, -0.0134, 0.0186,
-0.0332, -0.0004, -0.0015, -0.0188, 0.0408, -0.0054, 0.0266,
-0.0853, 0.0345, -0.0628, 0.0126, -0.0216, -0.0097, -0.0415,
-0.0183, -0.0099, -0.0415, -0.0173, 0.0367, 0.0062, -0.0098,
-0.0209, 0.0221, 0.0008, -0.0790, -0.0071, -0.0648, 0.0196,
-0.0063, -0.0658, -0.0023, 0.0247, -0.0106, 0.0212, -0.0085], device='cuda:0'))])
(IMPLEMENTATION) Test the Model¶
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.
In [79]:
def test(loaders, model, criterion, use_cuda):
# monitor test loss and accuracy
test_loss = 0.
correct = 0.
total = 0.
model.eval()
for batch_idx, (data, target) in enumerate(loaders['test']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# forward pass: compute predicted outputs by passing inputs to the model
output = model(data)
# calculate the loss
loss = criterion(output, target)
# update average test loss
test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
# convert output probabilities to predicted class
pred = output.data.max(1, keepdim=True)[1]
# compare predictions to true label
correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
total += data.size(0)
print('Test Loss: {:.6f}\n'.format(test_loss))
print('\nTest Accuracy: %2d%% (%2d/%2d)' % (
100. * correct / total, correct, total))
# call test function
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
Test Loss: 3.705375 Test Accuracy: 13% (117/836)
Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)¶
You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.
(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset¶
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).
If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.
In [95]:
## TODO: Specify data loaders
loaders_transfer = {
'train': trainloader,
'valid': validloader,
'test': testloader
}
As in dog detetction with fully pretrained model ,resnet 50 performed better than vgg , we'll be using RESNET 50¶
(IMPLEMENTATION) Model Architecture¶
Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.
In [96]:
# Load the pretrained model from pytorch
res50 = models.resnet50(pretrained=True)
# print out the model structure
print(res50)
ResNet(
(conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
(layer1): Sequential(
(0): Bottleneck(
(conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
(layer2): Sequential(
(0): Bottleneck(
(conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(3): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
(layer3): Sequential(
(0): Bottleneck(
(conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(3): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(4): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(5): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
(layer4): Sequential(
(0): Bottleneck(
(conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
(avgpool): AvgPool2d(kernel_size=7, stride=1, padding=0)
(fc): Linear(in_features=2048, out_features=1000, bias=True)
)
In [97]:
print(res50.fc.out_features)
1000
In [98]:
for name,child in res50.named_children():
print(name)
print('**',child)
print('---')
conv1
** Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
---
bn1
** BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
---
relu
** ReLU(inplace)
---
maxpool
** MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
---
layer1
** Sequential(
(0): Bottleneck(
(conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
---
layer2
** Sequential(
(0): Bottleneck(
(conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(3): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
---
layer3
** Sequential(
(0): Bottleneck(
(conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(3): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(4): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(5): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
---
layer4
** Sequential(
(0): Bottleneck(
(conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
---
avgpool
** AvgPool2d(kernel_size=7, stride=1, padding=0)
---
fc
** Linear(in_features=2048, out_features=1000, bias=True)
---
In [99]:
import torchvision.models as models
import torch.nn as nn
model_transfer=res50
for name,child in model_transfer.named_children():
if name in ['layer4','fc']:
print(name + 'is unfrozen')
for param in child.parameters():
param.requires_grad = True
else:
print(name + 'is frozen')
for param in child.parameters():
param.requires_grad = False
model_transfer.fc = nn.Sequential(
nn.Linear(2048, 516),
nn.ReLU(inplace=True),
nn.Linear(516, 133))
if use_cuda:
model_transfer = model_transfer.cuda()
conv1is frozen bn1is frozen reluis frozen maxpoolis frozen layer1is frozen layer2is frozen layer3is frozen layer4is unfrozen avgpoolis frozen fcis unfrozen
In [100]:
model_transfer
Out[100]:
ResNet(
(conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
(layer1): Sequential(
(0): Bottleneck(
(conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
(layer2): Sequential(
(0): Bottleneck(
(conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(3): Bottleneck(
(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
(layer3): Sequential(
(0): Bottleneck(
(conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(3): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(4): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(5): Bottleneck(
(conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
(layer4): Sequential(
(0): Bottleneck(
(conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
(downsample): Sequential(
(0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Bottleneck(
(conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
(2): Bottleneck(
(conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace)
)
)
(avgpool): AvgPool2d(kernel_size=7, stride=1, padding=0)
(fc): Sequential(
(0): Linear(in_features=2048, out_features=516, bias=True)
(1): ReLU(inplace)
(2): Linear(in_features=516, out_features=133, bias=True)
)
)
Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.
Answer:
- Chose resnet50 beacuse in dog face detetction with all param freezed , it performed better and also
- light weight model
- higher accuracy
Residual networks is based on theory that as we go deeper, we should make sure not to degrade accuracy and so, Keep learning the residuals to match the predicted with the actual and this is acheived by mapping the identity function.
This architechture contains
- (conv1) as first convolutional layer containing in channels as 3 which is due to RGB input tensor
- (bn1) as batch normalization layer
- followed by ReLU and MaxPooling
- then it contains 4 main layers named layer1, layer2, layer3 and layer4 which contains further sub layers of convolution
- followed by batchnorm
- followed by relu
- followed by maxpooling
- and then finally fc.
ReLU activation is used as it's the most proven activation function for classification problems as it introduces good and right amount of non linearity with less chances of vanishing gradient problem ! Batch normalization helped in making the network more stable and learning faster thereby faster convergence. Maxpooling helped in downsampling high number of parameters created by producing higher dimensional feature maps after convolution operation and thus selecting only relevant features from the high dimensioned feature matrix. Then i replaced last layer of this architechture by fully connected layer containing two sub linear layers as follows : Linear(in_features=2048, out_features=512) Linear(in_features=512, out_features=133) with ReLU activations between the linears. Now, the important thing is about unfrozing layers, so i choose to first only train layer4 and fc and then i unfroze one more layer layer3 which increased the accuracy and decreased the validation losses.
(IMPLEMENTATION) Specify Loss Function and Optimizer¶
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.
In [101]:
criterion_transfer = nn.CrossEntropyLoss()
optimizer_transfer = optim.SGD(filter(lambda p:p.requires_grad,model_transfer.parameters()), lr=0.002, momentum=0.9)
## need to pass only those param that are unfrrozen!!
(IMPLEMENTATION) Train and Validate the Model¶
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.
In [102]:
# train the model
n_epochs=50
model_transfer = train(n_epochs, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')
# load the model that got the best validation accuracy (uncomment the line below)
Epoch: 1 Training Loss: 4.722241 Validation Loss: 4.260810 Validation loss decreased (inf --> 4.260810). Saving model ... Epoch: 2 Training Loss: 3.891512 Validation Loss: 2.719768 Validation loss decreased (4.260810 --> 2.719768). Saving model ... Epoch: 3 Training Loss: 2.648591 Validation Loss: 1.566293 Validation loss decreased (2.719768 --> 1.566293). Saving model ... Epoch: 4 Training Loss: 1.821002 Validation Loss: 1.014901 Validation loss decreased (1.566293 --> 1.014901). Saving model ... Epoch: 5 Training Loss: 1.387854 Validation Loss: 0.795051 Validation loss decreased (1.014901 --> 0.795051). Saving model ... Epoch: 6 Training Loss: 1.167945 Validation Loss: 0.748533 Validation loss decreased (0.795051 --> 0.748533). Saving model ... Epoch: 7 Training Loss: 0.997539 Validation Loss: 0.564847 Validation loss decreased (0.748533 --> 0.564847). Saving model ... Epoch: 8 Training Loss: 0.923837 Validation Loss: 0.496929 Validation loss decreased (0.564847 --> 0.496929). Saving model ... Epoch: 9 Training Loss: 0.846039 Validation Loss: 0.471830 Validation loss decreased (0.496929 --> 0.471830). Saving model ... Epoch: 10 Training Loss: 0.810553 Validation Loss: 0.438906 Validation loss decreased (0.471830 --> 0.438906). Saving model ... Epoch: 11 Training Loss: 0.750848 Validation Loss: 0.430277 Validation loss decreased (0.438906 --> 0.430277). Saving model ... Epoch: 12 Training Loss: 0.711109 Validation Loss: 0.447217 Epoch: 13 Training Loss: 0.683180 Validation Loss: 0.384616 Validation loss decreased (0.430277 --> 0.384616). Saving model ... Epoch: 14 Training Loss: 0.642066 Validation Loss: 0.360080 Validation loss decreased (0.384616 --> 0.360080). Saving model ... Epoch: 15 Training Loss: 0.646684 Validation Loss: 0.345113 Validation loss decreased (0.360080 --> 0.345113). Saving model ... Epoch: 16 Training Loss: 0.606383 Validation Loss: 0.345510 Epoch: 17 Training Loss: 0.591615 Validation Loss: 0.411103 Epoch: 18 Training Loss: 0.546986 Validation Loss: 0.405461 Epoch: 19 Training Loss: 0.529621 Validation Loss: 0.342638 Validation loss decreased (0.345113 --> 0.342638). Saving model ... Epoch: 20 Training Loss: 0.520297 Validation Loss: 0.336307 Validation loss decreased (0.342638 --> 0.336307). Saving model ... Epoch: 21 Training Loss: 0.511053 Validation Loss: 0.376479 Epoch: 22 Training Loss: 0.503074 Validation Loss: 0.308788 Validation loss decreased (0.336307 --> 0.308788). Saving model ... Epoch: 23 Training Loss: 0.482304 Validation Loss: 0.353287 Epoch: 24 Training Loss: 0.467420 Validation Loss: 0.319643 Epoch: 25 Training Loss: 0.466349 Validation Loss: 0.337898 Epoch: 26 Training Loss: 0.446002 Validation Loss: 0.301182 Validation loss decreased (0.308788 --> 0.301182). Saving model ... Epoch: 27 Training Loss: 0.457967 Validation Loss: 0.328277 Epoch: 28 Training Loss: 0.429792 Validation Loss: 0.308517 Epoch: 29 Training Loss: 0.385724 Validation Loss: 0.331309 Epoch: 30 Training Loss: 0.428134 Validation Loss: 0.331316 Epoch: 31 Training Loss: 0.413687 Validation Loss: 0.342136 Epoch: 32 Training Loss: 0.411828 Validation Loss: 0.299753 Validation loss decreased (0.301182 --> 0.299753). Saving model ... Epoch: 33 Training Loss: 0.401343 Validation Loss: 0.291624 Validation loss decreased (0.299753 --> 0.291624). Saving model ... Epoch: 34 Training Loss: 0.387029 Validation Loss: 0.307140 Epoch: 35 Training Loss: 0.357279 Validation Loss: 0.312482 Epoch: 36 Training Loss: 0.388508 Validation Loss: 0.294033 Epoch: 37 Training Loss: 0.363480 Validation Loss: 0.338502 Epoch: 38 Training Loss: 0.371871 Validation Loss: 0.340184 Epoch: 39 Training Loss: 0.349409 Validation Loss: 0.316996 Epoch: 40 Training Loss: 0.337181 Validation Loss: 0.291183 Validation loss decreased (0.291624 --> 0.291183). Saving model ... Epoch: 41 Training Loss: 0.349908 Validation Loss: 0.347648 Epoch: 42 Training Loss: 0.354147 Validation Loss: 0.329109 Epoch: 43 Training Loss: 0.325253 Validation Loss: 0.312263 Epoch: 44 Training Loss: 0.329482 Validation Loss: 0.395118 Epoch: 45 Training Loss: 0.327995 Validation Loss: 0.341807 Epoch: 46 Training Loss: 0.334716 Validation Loss: 0.348989 Epoch: 47 Training Loss: 0.318253 Validation Loss: 0.341828 Epoch: 48 Training Loss: 0.295617 Validation Loss: 0.305116 Epoch: 49 Training Loss: 0.305102 Validation Loss: 0.329744 Epoch: 50 Training Loss: 0.321788 Validation Loss: 0.332749
In [104]:
model_transfer.load_state_dict(torch.load('model_transfer.pt'))
In [113]:
y=torch.load('model_transfer.pt')
In [114]:
print(y)
OrderedDict([('conv1.weight', tensor([[[[ 1.3335e-02, 1.4664e-02, -1.5351e-02, ..., -4.0896e-02,
-4.3034e-02, -7.0755e-02],
[ 4.1205e-03, 5.8477e-03, 1.4948e-02, ..., 2.2060e-03,
-2.0912e-02, -3.8517e-02],
[ 2.2331e-02, 2.3595e-02, 1.6120e-02, ..., 1.0281e-01,
6.2641e-02, 5.1977e-02],
...,
[-9.0349e-04, 2.7767e-02, -1.0105e-02, ..., -1.2722e-01,
-7.6604e-02, 7.8453e-03],
[ 3.5894e-03, 4.8006e-02, 6.2051e-02, ..., 2.4267e-02,
-3.3662e-02, -1.5709e-02],
[-8.0029e-02, -3.2238e-02, -1.7808e-02, ..., 3.5359e-02,
2.2439e-02, 1.7077e-03]],
[[-1.8452e-02, 1.1415e-02, 2.3850e-02, ..., 5.3736e-02,
4.4022e-02, -9.4675e-03],
[-7.7273e-03, 1.8890e-02, 6.7981e-02, ..., 1.5956e-01,
1.4606e-01, 1.1999e-01],
[-4.6013e-02, -7.6075e-02, -8.9648e-02, ..., 1.2108e-01,
1.6705e-01, 1.7619e-01],
...,
[ 2.8818e-02, 1.3665e-02, -8.3825e-02, ..., -3.8081e-01,
-3.0414e-01, -1.3966e-01],
[ 8.2868e-02, 1.3864e-01, 1.5241e-01, ..., -5.1232e-03,
-1.2435e-01, -1.2967e-01],
[-7.2789e-03, 7.7021e-02, 1.3999e-01, ..., 1.8427e-01,
1.1144e-01, 2.3438e-02]],
[[-1.8311e-02, -5.6424e-03, 8.7224e-03, ..., 2.5775e-02,
2.6431e-02, -3.9914e-03],
[-1.0098e-02, 4.1615e-03, 4.9730e-02, ..., 1.2447e-01,
1.1950e-01, 1.1198e-01],
[-6.3478e-02, -1.0146e-01, -9.8343e-02, ..., 1.0630e-01,
1.3982e-01, 1.4942e-01],
...,
[ 2.5810e-02, 1.0501e-02, -7.4578e-02, ..., -3.1385e-01,
-2.5495e-01, -1.2276e-01],
[ 7.3049e-02, 1.1170e-01, 1.3093e-01, ..., -6.4584e-03,
-1.2548e-01, -1.2446e-01],
[-6.4210e-03, 6.6393e-02, 1.2177e-01, ..., 1.9075e-01,
1.1415e-01, 2.3337e-02]]],
[[[ 6.8609e-02, 3.7955e-02, 5.3564e-02, ..., 2.6891e-02,
4.8369e-02, 6.3264e-02],
[ 6.1844e-02, 1.8407e-02, 2.2672e-02, ..., -4.8800e-02,
-2.2130e-02, -5.7287e-03],
[ 5.6553e-02, 1.4883e-02, -6.9185e-03, ..., -1.2919e-01,
-9.5042e-02, -5.8671e-02],
...,
[ 2.3802e-02, -5.2273e-02, -1.1277e-01, ..., -2.5591e-01,
-2.4049e-01, -2.0315e-01],
[ 5.6221e-02, -2.1824e-02, -5.9207e-02, ..., -2.3806e-01,
-1.9836e-01, -1.6578e-01],
[ 5.9635e-02, 3.7172e-03, -4.8716e-02, ..., -1.6098e-01,
-1.4336e-01, -1.0251e-01]],
[[-9.9041e-02, -7.2129e-02, -7.2748e-02, ..., -3.6232e-02,
-8.1876e-02, -8.8119e-02],
[-7.0621e-02, -3.9200e-02, -1.0514e-02, ..., 5.9519e-02,
2.5891e-02, -1.3287e-02],
[-9.3963e-02, -2.5318e-02, 3.0725e-02, ..., 2.0738e-01,
1.6162e-01, 8.6865e-02],
...,
[-3.9978e-02, 6.4148e-02, 1.6941e-01, ..., 4.5814e-01,
3.7839e-01, 2.5870e-01],
[-6.4985e-02, 1.3637e-02, 1.3008e-01, ..., 3.6661e-01,
3.2009e-01, 2.0442e-01],
[-1.0433e-01, -2.7216e-02, 4.1199e-02, ..., 2.2797e-01,
1.8622e-01, 1.1854e-01]],
[[ 4.2717e-02, 4.7269e-02, 1.7798e-02, ..., 3.4693e-02,
2.8595e-02, 4.3173e-02],
[ 3.6308e-02, 3.3459e-02, -1.2449e-03, ..., 4.2910e-03,
4.6513e-03, 2.4962e-02],
[ 2.3541e-02, 1.6826e-02, -1.5325e-02, ..., -7.2865e-02,
-7.2030e-02, -2.3747e-02],
...,
[ 2.9623e-02, 1.8176e-03, -7.9559e-02, ..., -1.8138e-01,
-1.6950e-01, -6.1192e-02],
[ 3.5233e-02, -8.5295e-03, -5.4253e-02, ..., -1.5068e-01,
-1.2747e-01, -4.5340e-02],
[ 5.4786e-02, 3.9805e-02, -1.0473e-02, ..., -4.2332e-02,
-5.2220e-02, 5.1277e-04]]],
[[[ 7.3052e-03, 1.5686e-02, -2.4423e-04, ..., -3.2539e-02,
-3.1297e-02, -3.0249e-02],
[ 8.6377e-03, 2.4622e-02, 1.1249e-02, ..., -1.0645e-02,
-1.5026e-02, -3.4091e-02],
[ 7.7960e-03, 3.1337e-02, 2.5109e-02, ..., 1.2514e-02,
-5.7764e-04, -5.3505e-03],
...,
[-1.0597e-03, 1.3395e-02, 1.1477e-03, ..., 3.6588e-02,
4.6791e-02, 7.5218e-02],
[-1.5273e-02, 8.1007e-03, -7.3382e-04, ..., 3.7322e-03,
1.9803e-02, 7.4966e-02],
[-1.9642e-02, -1.8543e-03, -6.9084e-03, ..., -3.2996e-03,
2.8972e-02, 8.7559e-02]],
[[ 1.2775e-02, 8.8500e-03, 2.3261e-03, ..., 1.3541e-02,
3.6049e-02, 4.8399e-02],
[ 1.4153e-02, 1.1684e-02, -2.9641e-03, ..., 2.7169e-02,
5.1851e-02, 3.8690e-02],
[ 3.1984e-03, 5.0850e-03, -9.3271e-03, ..., 4.1137e-02,
5.4548e-02, 4.5094e-02],
...,
[-1.7035e-02, -4.2867e-02, -8.4639e-02, ..., -4.9248e-02,
-2.3901e-02, 2.6232e-03],
[-2.8155e-03, -8.8370e-03, -4.8081e-02, ..., -8.6432e-02,
-7.1647e-02, -2.3398e-02],
[ 2.8788e-02, 2.3000e-02, -4.6879e-03, ..., -5.0775e-02,
-3.9457e-02, -7.1134e-03]],
[[ 9.3511e-03, -3.6039e-02, -3.1732e-02, ..., 3.4844e-03,
4.3197e-02, 4.5083e-02],
[-1.1875e-02, -5.8264e-02, -6.3004e-02, ..., -1.7674e-02,
1.5884e-02, -7.6554e-03],
[ 1.4215e-02, -3.3805e-02, -5.0096e-02, ..., -2.1366e-03,
6.0525e-03, -8.0126e-03],
...,
[ 8.0828e-03, -4.9916e-02, -9.0253e-02, ..., -8.7565e-02,
-9.9468e-02, -1.0924e-01],
[ 1.8781e-02, -1.4367e-02, -4.6708e-02, ..., -1.1187e-01,
-1.4514e-01, -1.4444e-01],
[ 5.7803e-02, 1.7252e-02, -5.7767e-03, ..., -7.4414e-02,
-1.1313e-01, -1.2434e-01]]],
...,
[[[ 1.5639e-02, 2.0411e-02, 3.5717e-02, ..., 7.1576e-03,
4.5493e-02, -7.6534e-03],
[-2.1845e-02, 5.6129e-02, 6.0663e-02, ..., -5.6907e-02,
8.5169e-02, -2.6663e-02],
[-4.7579e-02, 1.0689e-01, 7.5667e-02, ..., -8.8005e-02,
1.5159e-01, -3.2074e-02],
...,
[ 1.4774e-03, 1.2554e-01, -4.9284e-02, ..., 5.9099e-02,
1.5756e-01, -4.1167e-02],
[ 5.0183e-03, 6.2567e-02, -5.8753e-02, ..., 4.7302e-02,
8.2366e-02, -4.6016e-02],
[ 1.6201e-02, 5.4047e-03, -6.5849e-02, ..., 5.1001e-02,
4.6469e-02, -1.6759e-02]],
[[ 1.0200e-02, 4.4592e-02, 1.5321e-03, ..., -6.6545e-02,
4.5731e-02, 4.8529e-02],
[ 4.5088e-03, 1.2873e-01, 3.0149e-02, ..., -1.4642e-01,
1.4743e-01, 8.3739e-02],
[ 5.6123e-03, 2.1745e-01, 3.0934e-02, ..., -1.8033e-01,
2.6112e-01, 9.4205e-02],
...,
[ 4.6450e-02, 1.9461e-01, -1.3893e-01, ..., 4.3828e-02,
2.8577e-01, 5.1682e-02],
[ 2.5445e-02, 9.4402e-02, -1.3597e-01, ..., 5.5534e-02,
1.5707e-01, -1.3938e-02],
[ 1.6235e-02, 2.9242e-02, -1.0514e-01, ..., 7.2561e-02,
8.2445e-02, -1.0449e-02]],
[[ 7.6312e-04, 2.2041e-02, 1.2029e-02, ..., -3.9314e-02,
7.8765e-03, 1.1339e-02],
[-2.6744e-02, 7.6046e-02, 6.5150e-02, ..., -5.3509e-02,
7.7586e-02, 1.5460e-02],
[-4.3151e-02, 1.2443e-01, 7.8582e-02, ..., -7.2363e-02,
1.3668e-01, -3.7664e-03],
...,
[-4.8974e-03, 1.2171e-01, -4.9993e-02, ..., 4.0816e-02,
1.2784e-01, -2.7264e-02],
[-5.7832e-03, 6.6111e-02, -4.9786e-02, ..., 3.8487e-02,
6.9511e-02, -4.3904e-02],
[ 3.3817e-03, 3.2986e-02, -4.2698e-02, ..., 4.3439e-02,
3.3467e-02, -2.9978e-02]]],
[[[ 4.5842e-02, 5.2311e-02, 4.4927e-02, ..., -2.9410e-02,
4.5506e-03, 1.4374e-02],
[ 5.2478e-02, 5.1215e-02, 4.7960e-02, ..., -1.1274e-01,
-8.2460e-02, -2.5520e-02],
[ 9.0280e-02, 7.7345e-02, 6.7258e-02, ..., -2.1452e-01,
-1.1146e-01, -1.7177e-02],
...,
[ 3.2732e-02, 7.1492e-05, -1.5341e-01, ..., -2.5537e-01,
-1.1447e-01, 4.2084e-02],
[ 1.9229e-02, -2.7870e-02, -1.4789e-01, ..., -2.4871e-01,
-3.0199e-02, 8.2352e-02],
[-6.6101e-03, -5.7227e-02, -1.5145e-01, ..., -1.8681e-01,
9.7780e-03, 9.8369e-02]],
[[-3.1084e-03, 1.2002e-02, 1.6918e-02, ..., 4.7799e-03,
4.7867e-03, -8.5661e-04],
[ 1.9549e-02, 1.5186e-02, 3.4710e-02, ..., -4.2644e-03,
-1.3299e-02, -5.1674e-03],
[ 2.0221e-02, 1.1181e-02, 5.0552e-02, ..., -6.5437e-02,
-6.4139e-03, 1.8563e-02],
...,
[-1.1129e-02, 1.0629e-02, -4.2425e-02, ..., -4.8182e-02,
-2.8246e-02, 4.6424e-02],
[ 7.5653e-03, 1.1309e-02, -1.3535e-02, ..., -6.6639e-02,
1.0465e-02, 3.6264e-02],
[ 9.8492e-03, 6.3770e-03, -1.5771e-04, ..., -3.0868e-02,
2.6524e-02, 3.0090e-02]],
[[-4.2418e-02, -2.3199e-02, -2.2478e-02, ..., 1.4661e-02,
1.2153e-02, -2.1124e-02],
[-2.9424e-02, -2.5660e-02, 1.4858e-03, ..., 4.2752e-02,
2.1531e-02, -5.6081e-03],
[-3.9465e-02, -4.5056e-02, 1.4975e-02, ..., 3.0639e-02,
4.1439e-02, -7.7498e-03],
...,
[-3.1252e-02, 7.7716e-03, 1.5953e-02, ..., 6.3187e-02,
4.8559e-03, -2.9812e-02],
[-1.2930e-02, 1.6789e-02, 4.8435e-02, ..., 4.6301e-02,
2.6589e-02, -4.1627e-02],
[-1.7138e-02, 7.2573e-03, 4.8229e-02, ..., 3.5378e-02,
1.4400e-02, -4.7261e-02]]],
[[[ 1.9836e-02, -2.8316e-02, 6.1643e-02, ..., -5.2705e-02,
9.8331e-03, 4.9678e-03],
[-4.6186e-02, 5.3722e-02, -2.6572e-02, ..., 1.9153e-02,
-2.2031e-02, -1.3209e-02],
[ 5.2939e-02, -4.4223e-02, -3.7457e-02, ..., -2.1368e-01,
-2.1620e-03, 3.4433e-02],
...,
[-3.5689e-02, 9.3666e-02, 1.1909e-01, ..., 5.2430e-01,
-4.5373e-02, -5.9195e-02],
[-1.1405e-02, 1.0079e-01, -1.9849e-01, ..., 2.3351e-01,
-2.6296e-01, 1.2434e-01],
[-1.4071e-02, 5.2204e-03, -7.5740e-02, ..., -5.6693e-02,
3.3713e-02, -3.4239e-02]],
[[ 1.8626e-02, -3.9080e-02, 4.1489e-02, ..., -4.2309e-02,
2.7354e-02, -5.3296e-03],
[-4.5311e-02, 6.2526e-02, -1.8939e-02, ..., 2.9117e-02,
3.0331e-03, -1.8276e-03],
[ 3.6988e-02, -6.5929e-02, -2.1645e-02, ..., -2.6028e-01,
5.6034e-02, 3.7797e-02],
...,
[-3.4665e-02, 1.2999e-01, 1.2347e-01, ..., 6.5182e-01,
-1.6673e-02, -1.2347e-01],
[-1.9596e-02, 1.2236e-01, -2.4884e-01, ..., 3.3043e-01,
-3.2030e-01, 9.4385e-02],
[ 2.0938e-03, 2.5218e-02, -8.2988e-02, ..., -6.9019e-02,
2.0379e-02, -1.5437e-02]],
[[ 1.3068e-02, -3.7082e-02, 1.9962e-02, ..., -4.6411e-02,
2.6141e-02, 2.5953e-03],
[-3.3303e-02, 7.5309e-02, -2.8833e-02, ..., 4.7499e-02,
-7.2355e-03, -1.5955e-02],
[ 6.1195e-02, -4.4915e-02, -1.3650e-01, ..., -1.3014e-01,
6.3804e-03, 1.3756e-02],
...,
[-1.5692e-02, 8.1412e-03, 1.4574e-01, ..., 4.6457e-01,
-1.6762e-01, -2.8489e-02],
[-4.5328e-02, 3.7344e-02, -1.2535e-01, ..., 1.0561e-01,
-2.7718e-01, 1.6692e-01],
[ 1.0456e-02, 1.9020e-02, -1.5351e-02, ..., -1.1350e-01,
6.7615e-02, -6.7650e-03]]]], device='cuda:0')), ('bn1.weight', tensor([ 2.3888e-01, 2.9136e-01, 3.1615e-01, 2.7122e-01, 2.1731e-01,
3.0903e-01, 2.2937e-01, 2.3086e-01, 2.1129e-01, 2.8054e-01,
1.9923e-01, 3.1894e-01, 1.7955e-01, 1.1246e-08, 1.9704e-01,
2.0996e-01, 2.4317e-01, 2.1697e-01, 1.9415e-01, 3.1569e-01,
1.9648e-01, 2.3214e-01, 2.1962e-01, 2.1633e-01, 2.4357e-01,
2.9683e-01, 2.3852e-01, 2.1162e-01, 1.4492e-01, 2.9388e-01,
2.2911e-01, 9.2716e-02, 4.3334e-01, 2.0782e-01, 2.7990e-01,
3.5804e-01, 2.9315e-01, 2.5306e-01, 2.4210e-01, 2.1755e-01,
3.8645e-01, 2.1003e-01, 3.6805e-01, 3.3724e-01, 5.0826e-01,
1.9341e-01, 2.3914e-01, 2.6652e-01, 3.9020e-01, 1.9840e-01,
2.1694e-01, 2.6666e-01, 4.9806e-01, 2.3553e-01, 2.1349e-01,
2.5951e-01, 2.3547e-01, 1.7579e-01, 4.5354e-01, 1.7102e-01,
2.4903e-01, 2.5148e-01, 3.8020e-01, 1.9665e-01], device='cuda:0')), ('bn1.bias', tensor([ 2.2484e-01, 6.0617e-01, 1.2483e-02, 1.3270e-01, 1.8030e-01,
1.4739e-01, 1.7430e-01, 1.9023e-01, 2.3226e-01, 2.0082e-01,
1.2834e-01, -2.1285e-01, 1.5065e-01, -3.9217e-08, 2.4985e-01,
2.0454e-01, 5.4934e-01, 2.1021e-01, 2.2505e-01, 4.6484e-01,
2.3888e-01, 2.0442e-01, 2.1546e-01, 6.6194e-01, 2.2755e-01,
6.6069e-01, 2.0587e-01, 1.9292e-01, 1.1195e-01, 3.3785e-01,
1.2393e-01, 4.1079e-02, 7.7150e-01, 2.6964e-01, 3.3347e-01,
5.7908e-01, 1.5026e-01, 1.7534e-01, 1.9429e-01, 1.7248e-01,
8.0577e-01, 2.3693e-01, -4.3369e-01, 8.4813e-01, -3.7857e-01,
2.4787e-01, 1.8101e-01, 3.2949e-01, -2.8598e-01, 2.2717e-01,
2.6168e-01, 5.7609e-02, -5.0320e-01, 1.5704e-01, 1.7890e-01,
2.8114e-01, 4.2167e-01, -9.7650e-02, -3.1231e-01, -2.5637e-02,
8.8566e-02, 1.8052e-01, 8.3045e-01, 2.5015e-01], device='cuda:0')), ('bn1.running_mean', tensor([ 1.0809e-02, 1.7091e-02, 1.1719e-01, -1.9685e-03, -1.7913e-03,
-3.4613e-05, 5.6143e-03, 7.1892e-03, 3.2146e-03, 1.1383e-02,
3.1447e-03, 5.1101e-02, -1.5220e-02, -2.9003e-08, 3.2541e-03,
3.5492e-03, -3.2863e-02, -8.3364e-03, -3.9215e-04, 2.7882e-01,
-7.1441e-04, -1.6779e-03, -5.1888e-03, -4.4259e-02, -3.8837e-03,
4.2526e-02, -1.2899e-03, 7.4543e-03, 7.2724e-03, -2.1655e-01,
-6.4606e-03, 1.7388e-04, -1.3842e-01, -2.7626e-03, -1.9950e-01,
1.4827e-01, -8.7787e-05, -4.4378e-03, -8.5877e-04, 1.3972e-03,
7.4684e-02, -2.0159e-02, 3.1877e-02, 1.0953e-01, 1.2665e-01,
-2.6964e-02, -2.4448e-03, -5.2884e-03, -1.0573e-02, -4.8945e-03,
1.8169e-03, -4.2912e-03, -2.2397e-02, -5.3786e-03, 1.1122e-03,
-8.4138e-03, -6.0905e-02, -4.2915e-02, -8.6269e-04, 4.3075e-03,
-1.4891e-03, 1.3860e-03, -6.4907e-02, -2.2993e-03], device='cuda:0')), ('bn1.running_var', tensor([ 7.6342e-01, 2.7563e+00, 3.9667e+00, 3.8788e+00, 4.2595e-01,
6.5522e+00, 5.1682e-01, 4.2523e-01, 3.0386e-01, 3.5791e+00,
2.5820e-01, 2.3643e+00, 1.8563e-01, 2.8209e-12, 9.1606e-02,
3.6167e-01, 1.2930e+00, 4.0843e-01, 1.4057e-01, 5.6759e+00,
1.6268e-01, 3.9279e-01, 2.1921e-01, 1.9772e+00, 7.9053e-01,
3.6708e+00, 5.3803e-01, 4.0465e-01, 4.9939e-02, 3.8859e+00,
9.7682e-01, 4.1449e-02, 1.2856e+01, 8.0791e-02, 4.0289e+00,
7.3452e+00, 5.6830e+00, 2.2289e+00, 5.3521e-01, 5.7085e-01,
9.6661e+00, 1.1343e-01, 2.6222e+00, 4.4371e+00, 8.1324e+00,
2.8851e-01, 4.7489e-01, 2.0408e+00, 1.4719e+00, 2.0916e-01,
3.9689e-01, 1.2497e+00, 4.1890e+00, 4.5221e-01, 5.4469e-01,
1.7006e+00, 1.0680e+00, 2.2923e-01, 4.5978e+00, 3.7976e-01,
1.0428e+00, 1.0911e+00, 8.8885e+00, 1.5466e-01], device='cuda:0')), ('layer1.0.conv1.weight', tensor([[[[ 3.5144e-03]],
[[ 3.9855e-02]],
[[-2.4795e-02]],
...,
[[ 1.0283e-01]],
[[-2.8931e-01]],
[[-1.5175e-01]]],
[[[ 1.1493e-01]],
[[ 5.6982e-03]],
[[ 2.8378e-02]],
...,
[[ 8.0681e-02]],
[[-8.4798e-02]],
[[-1.3576e-01]]],
[[[-1.9008e-02]],
[[ 1.1191e-02]],
[[-1.6003e-02]],
...,
[[ 7.1472e-02]],
[[ 7.6694e-02]],
[[-4.6049e-02]]],
...,
[[[-1.9799e-03]],
[[-1.9214e-03]],
[[ 2.3714e-02]],
...,
[[ 1.6071e-02]],
[[-6.2259e-02]],
[[ 9.3809e-03]]],
[[[-5.2812e-02]],
[[ 1.0688e-02]],
[[ 3.2017e-02]],
...,
[[-2.1943e-02]],
[[-6.1035e-03]],
[[ 1.2559e-02]]],
[[[-5.6478e-02]],
[[-1.1904e-01]],
[[ 4.7069e-02]],
...,
[[ 4.1034e-02]],
[[ 2.4489e-01]],
[[-7.9294e-03]]]], device='cuda:0')), ('layer1.0.bn1.weight', tensor([ 2.1341e-01, 1.8848e-01, 1.4136e-01, 1.5273e-01, 1.3220e-01,
1.8735e-01, 1.4475e-01, 4.5110e-08, 1.5993e-01, 1.4946e-01,
2.3499e-01, 1.8315e-01, 1.8516e-01, 1.4933e-01, 1.3090e-01,
1.0634e-01, 3.7487e-01, 1.2644e-01, 3.1895e-01, 2.7160e-01,
2.5810e-01, 2.9458e-01, 1.8395e-01, 2.1088e-08, 3.3313e-01,
2.0461e-01, 3.0399e-01, 1.1805e-08, 1.4977e-01, 1.5719e-01,
1.4011e-01, 1.4900e-01, 1.2438e-01, 1.8786e-01, 1.4257e-01,
3.4828e-01, 1.5038e-01, 3.0034e-01, 2.5925e-01, 1.0711e-01,
2.6875e-01, 1.3552e-01, 1.1822e-01, 1.1189e-01, 2.8736e-01,
3.2637e-01, 1.4781e-01, 2.3105e-01, 3.3638e-01, 2.8808e-01,
1.2319e-01, 3.0763e-01, 1.1846e-01, 1.3137e-01, 2.0671e-01,
1.5787e-01, 2.6574e-08, 2.0467e-01, 2.8797e-08, 1.8284e-01,
3.0180e-01, 1.7401e-01, 2.8438e-01, 2.3715e-01], device='cuda:0')), ('layer1.0.bn1.bias', tensor([ 4.3266e-01, 4.6854e-02, -8.0134e-02, 7.3302e-02, 2.7970e-01,
-7.8047e-03, 9.4087e-02, -1.0086e-07, -1.4034e-01, -5.1599e-02,
4.4470e-02, 2.1814e-01, 4.0718e-02, 1.1979e-01, 1.4432e-01,
1.3672e-01, -1.1168e-01, 1.4774e-01, -1.2879e-01, -5.3147e-02,
-3.3920e-02, -2.0600e-02, 6.2783e-02, -6.5736e-08, -7.1213e-02,
6.9510e-02, -1.3264e-01, -6.4411e-08, -2.8908e-02, 9.4164e-02,
2.4790e-01, -8.2850e-02, -2.8872e-02, -1.7086e-01, 9.9522e-02,
-1.1357e-01, 1.9770e-01, 1.4800e-02, -7.0896e-02, 1.0722e-01,
1.2536e-02, -3.6633e-02, 1.4959e-01, 1.0533e-01, 2.0933e-02,
-1.0502e-01, -4.8848e-02, 4.9007e-01, -1.4755e-01, -1.0900e-01,
1.9815e-02, -7.0964e-02, -4.6543e-02, 1.0874e-01, -2.7878e-01,
4.4500e-03, -7.7156e-08, 7.5060e-02, -8.4474e-08, 2.2533e-01,
-7.1593e-02, -1.5823e-01, -3.4459e-02, 5.2894e-01], device='cuda:0')), ('layer1.0.bn1.running_mean', tensor([-6.5209e-01, -4.2048e-01, 2.6633e-01, -2.8898e-01, -9.3108e-02,
-3.5313e-01, -1.1393e-01, -1.5442e-07, 2.1620e-01, -2.9387e-01,
-3.8006e-01, 1.2760e+00, -1.3684e-01, -2.2420e-01, -3.4403e-01,
-2.0778e-02, -4.3287e-01, 8.2101e-01, -3.9520e-01, -3.2645e-01,
-2.7599e-01, -1.8428e-01, -1.9180e-01, -1.3287e-08, -4.4901e-01,
-2.0712e-01, -3.1592e-01, 2.9930e-09, -2.6712e-01, -3.2643e-01,
-1.5797e-02, 2.2730e-01, 2.5271e-02, -1.4071e-01, -3.1431e-01,
-5.2575e-01, -2.4421e-01, -2.5573e-01, -3.3099e-01, 3.6533e-01,
-1.4551e-01, -1.2916e-01, -3.4192e-01, -4.9743e-01, -2.7101e-01,
-3.7717e-01, -4.2560e-01, -3.4291e-01, -2.6136e-01, -5.1326e-01,
-2.6590e-02, -3.4528e-01, 1.0116e-01, -3.6230e-01, -3.4743e-02,
-7.0487e-01, -1.1158e-08, -2.8045e-01, -6.9120e-08, 1.1664e+00,
-1.8912e-01, -8.5371e-03, -4.8222e-01, -1.2531e-01], device='cuda:0')), ('layer1.0.bn1.running_var', tensor([ 1.9399e-01, 5.0435e-02, 8.6365e-03, 1.6708e-02, 5.1078e-02,
1.8581e-02, 1.8542e-02, 3.3889e-15, 9.1892e-03, 1.4981e-02,
5.8381e-02, 1.5078e-01, 2.5721e-02, 1.5256e-02, 1.6109e-02,
7.8273e-03, 9.1612e-02, 6.2797e-02, 6.6138e-02, 6.9857e-02,
4.8749e-02, 9.5355e-02, 2.8504e-02, 6.7025e-16, 8.3277e-02,
2.7678e-02, 7.2523e-02, 5.6300e-16, 1.8188e-02, 2.5341e-02,
6.6066e-02, 9.3094e-03, 9.2300e-03, 1.7943e-02, 1.1648e-02,
1.0480e-01, 5.8855e-02, 8.5863e-02, 7.8183e-02, 2.0484e-02,
4.5323e-02, 1.5233e-02, 2.9083e-02, 8.2520e-03, 6.7919e-02,
8.0298e-02, 1.5091e-02, 2.0995e-01, 9.8022e-02, 7.2526e-02,
1.2035e-02, 6.9513e-02, 8.5598e-03, 1.6058e-02, 1.6162e-02,
3.1290e-02, 1.0233e-15, 3.0972e-02, 1.1813e-15, 1.1462e-01,
9.0234e-02, 1.8887e-02, 7.6855e-02, 2.5123e-01], device='cuda:0')), ('layer1.0.conv2.weight', tensor([[[[ 2.0919e-09, -1.4152e-09, 6.5850e-09],
[ 4.7922e-09, 3.2819e-09, 9.4284e-10],
[ 6.4763e-09, -3.0156e-09, -2.3958e-09]],
[[ 1.1471e-08, 1.2815e-08, 1.3568e-08],
[ 1.0737e-08, 4.9712e-09, 3.8516e-09],
[ 1.0572e-08, 1.2153e-08, 6.9175e-09]],
[[-6.8447e-09, -2.8470e-09, -1.5396e-09],
[ 4.8804e-09, -3.0935e-09, 2.3919e-09],
[ 5.2998e-09, -1.4133e-11, -7.5312e-09]],
...,
[[-6.4828e-09, -7.6425e-09, -9.5572e-09],
[-7.0080e-09, -4.0208e-09, -4.5819e-09],
[-4.9349e-09, -5.4604e-10, -1.3119e-08]],
[[-1.0458e-09, -6.9635e-09, -5.1090e-09],
[-3.4306e-10, -4.2656e-09, -2.2703e-09],
[-1.8570e-09, -4.8467e-09, -3.5571e-09]],
[[ 1.8746e-09, 8.6741e-10, -5.5741e-10],
[ 2.6233e-10, -5.0061e-09, 5.8350e-09],
[ 2.4322e-10, 1.6342e-10, 3.3284e-09]]],
[[[-2.3484e-02, 1.8706e-03, 2.5798e-02],
[ 3.5355e-03, -5.2646e-03, 1.9523e-02],
[ 2.4600e-02, 2.1933e-02, 1.6219e-04]],
[[-3.3788e-03, 3.2976e-03, 1.2707e-02],
[ 1.7570e-02, 1.5054e-02, 1.7517e-02],
[ 2.2276e-02, 2.2339e-02, -5.5213e-03]],
[[-3.8421e-03, -5.9462e-03, 1.5058e-02],
[-3.1875e-03, -2.5218e-02, -3.3736e-03],
[ 4.9239e-04, -1.9673e-02, -1.8729e-02]],
...,
[[ 5.2840e-03, 6.6714e-03, 1.1301e-02],
[ 1.3283e-02, 9.0130e-03, 6.2237e-03],
[ 2.1055e-02, 1.5775e-02, 8.4497e-03]],
[[ 5.9612e-04, 2.7906e-03, 6.5433e-03],
[ 8.4950e-03, 9.2682e-02, 5.1201e-03],
[-9.0937e-03, -6.4196e-03, -2.7363e-04]],
[[-2.0518e-02, -4.6551e-03, -2.9206e-02],
[-1.1651e-02, 1.6589e-02, -1.8225e-03],
[-3.0077e-02, 4.2854e-03, 3.0994e-03]]],
[[[ 3.8186e-02, 4.5909e-02, 2.6377e-02],
[ 4.5241e-02, 6.5042e-02, 5.4767e-02],
[ 4.5518e-02, 5.9040e-02, 3.6224e-02]],
[[ 6.4452e-03, -9.0585e-04, 7.1830e-03],
[ 6.7971e-03, 6.3635e-03, 2.1455e-02],
[ 1.1262e-02, 1.6089e-02, 1.7535e-02]],
[[ 4.0125e-03, -1.5436e-02, 6.1919e-03],
[-1.6548e-02, -3.7222e-03, 2.9918e-02],
[-4.1074e-03, -4.8143e-03, 1.5329e-02]],
...,
[[ 2.4917e-02, 1.9070e-02, 3.3504e-02],
[ 3.1252e-02, -7.5642e-03, 3.5888e-02],
[ 2.9346e-02, 3.3745e-02, 4.2878e-02]],
[[ 6.5392e-03, -1.6478e-02, 2.1843e-02],
[-1.3546e-02, 2.4295e-03, 6.9680e-03],
[ 3.1423e-03, -3.7818e-03, 3.2182e-03]],
[[ 7.3159e-03, -2.3604e-02, -1.1507e-02],
[-3.3326e-02, -4.4065e-02, -4.6069e-02],
[-3.1847e-02, -3.7215e-02, -2.4266e-02]]],
...,
[[[ 1.8211e-09, 2.9120e-09, -5.6310e-09],
[ 4.3212e-09, -1.8635e-09, -3.3057e-09],
[ 2.2608e-09, -1.2302e-09, 1.6536e-10]],
[[ 1.9862e-09, -2.7085e-09, -1.5189e-08],
[ 9.0284e-09, 2.6774e-09, -2.3783e-09],
[ 5.4744e-09, 4.7255e-09, -5.1706e-09]],
[[ 4.2245e-10, -6.2215e-09, -7.1381e-09],
[-4.5743e-09, -8.9245e-09, -4.1074e-09],
[-1.7179e-09, -1.4299e-10, -4.7585e-09]],
...,
[[-1.6792e-09, 3.7950e-09, 5.4831e-10],
[-4.9152e-09, -9.4884e-09, -6.3708e-09],
[ 8.2731e-09, 3.9463e-09, 3.8983e-09]],
[[ 4.0006e-09, -1.3585e-09, -2.7413e-12],
[-1.2756e-09, 2.5309e-09, -4.6894e-09],
[ 4.9880e-10, 3.3068e-09, -8.5953e-09]],
[[ 2.7820e-09, 7.1263e-10, 3.2218e-09],
[ 3.4826e-09, 5.6161e-09, -3.8766e-09],
[ 2.3460e-09, 6.0292e-09, 4.8710e-10]]],
[[[ 8.1531e-02, 5.8091e-02, 3.6932e-02],
[ 3.4071e-02, -1.3114e-02, -2.9585e-02],
[ 1.0524e-02, -3.7806e-02, -5.0393e-02]],
[[-7.4381e-03, -1.9210e-03, -3.3986e-03],
[ 1.3362e-02, 1.5364e-02, -1.0274e-03],
[-7.1327e-03, -2.5406e-02, -3.1194e-02]],
[[ 6.0968e-03, -1.9646e-02, -8.9450e-03],
[ 1.4864e-03, -1.2455e-02, 1.0394e-02],
[-2.6015e-03, -1.3369e-02, 1.5983e-02]],
...,
[[ 3.0871e-03, 6.2054e-04, -6.2765e-03],
[-4.0756e-03, -1.3461e-02, -1.4281e-02],
[-1.0118e-02, -2.1184e-02, -1.2876e-02]],
[[ 2.3335e-03, 1.8122e-02, 1.0758e-03],
[ 1.1701e-02, 2.8367e-02, 6.5168e-03],
[-1.4372e-02, 1.6207e-02, 1.3691e-02]],
[[-6.2637e-02, -2.5556e-02, 4.6860e-03],
[-3.8629e-02, 3.0089e-02, 4.4090e-02],
[ 1.2748e-03, 7.1193e-02, 6.3498e-02]]],
[[[-4.4771e-02, -3.8409e-02, 9.7917e-02],
[-1.7643e-01, 3.2383e-03, 2.8121e-01],
[-1.0757e-01, -4.9282e-02, 1.4446e-01]],
[[ 5.1254e-03, 1.9011e-03, 4.1052e-02],
[-3.4516e-02, -1.4285e-02, 2.9000e-02],
[-3.2176e-02, -3.1341e-03, 5.7282e-02]],
[[-3.8417e-02, -5.1169e-02, -2.7389e-02],
[-3.8442e-02, -1.0360e-01, -8.7211e-02],
[-4.4993e-02, -5.8818e-02, -3.8287e-02]],
...,
[[-2.1003e-02, -1.7952e-02, 6.6377e-03],
[-5.5188e-02, -1.3361e-03, 6.0603e-02],
[-1.4080e-02, 1.6523e-03, 2.5246e-02]],
[[ 2.2449e-03, 1.1214e-02, 1.3945e-02],
[-1.1470e-02, -6.9804e-03, 4.3370e-02],
[-1.2780e-02, 2.6532e-03, 6.3033e-02]],
[[ 1.9265e-01, -4.7934e-02, -1.4074e-01],
[ 3.7943e-01, -1.4166e-02, -3.0701e-01],
[ 2.4804e-01, -5.6059e-02, -1.9182e-01]]]], device='cuda:0')), ('layer1.0.bn2.weight', tensor([ 2.3224e-08, 1.3805e-01, 2.4155e-01, 1.3627e-01, 1.1577e-01,
1.8789e-01, 1.2005e-01, 1.3625e-01, 1.7837e-01, 1.3934e-01,
1.0924e-01, 1.7452e-01, 1.3525e-01, 1.5064e-01, 1.3912e-01,
1.2596e-01, 1.5728e-01, 1.3527e-01, 2.2781e-01, 4.6913e-07,
2.7175e-01, 1.6913e-01, 2.0877e-01, 2.0823e-01, 1.5117e-01,
1.7594e-01, 1.7099e-01, 1.8060e-01, 1.8968e-01, 1.8072e-01,
1.5526e-01, 1.3217e-01, 1.5328e-01, 2.2601e-01, 1.4130e-01,
1.4350e-01, 1.5354e-01, 1.6157e-01, 1.7243e-01, 2.2057e-01,
2.0561e-01, 2.2331e-01, 1.7635e-01, 1.1320e-01, 1.7612e-01,
2.0285e-01, 1.0247e-01, 2.6118e-01, 1.6209e-01, 1.4597e-01,
1.6170e-01, 1.7936e-01, 1.8842e-01, 1.2481e-01, 1.9059e-01,
1.7148e-01, 2.0484e-01, 1.3645e-01, 1.7423e-01, 1.5198e-03,
1.7775e-01, 9.5639e-09, 1.7993e-01, 2.3351e-01], device='cuda:0')), ('layer1.0.bn2.bias', tensor([-1.1646e-07, 3.8030e-02, 3.5188e-01, 1.2425e-01, 2.7166e-01,
3.4886e-01, 2.8938e-01, 1.5056e-01, -1.4002e-01, -8.9164e-02,
-2.8543e-02, 3.1789e-01, 2.8283e-01, -3.4367e-02, -1.3814e-01,
2.8288e-01, 3.7119e-01, -1.0956e-01, 2.2031e-01, -1.4170e-06,
8.4392e-02, 1.4088e-01, -4.2237e-02, 4.2857e-01, 1.0983e-01,
1.7426e-01, 1.6968e-01, -2.6071e-02, -5.6061e-02, -2.9518e-02,
7.3970e-02, 3.1024e-01, 3.4015e-01, 5.2495e-01, -2.8603e-03,
-9.6186e-02, 3.9447e-01, 1.6469e-01, 1.0085e-01, 1.0875e-01,
3.6777e-01, 7.9735e-02, 6.4225e-02, 2.2529e-01, -5.5575e-02,
-5.9820e-02, 2.7217e-01, -2.2807e-01, 3.8712e-01, -2.7749e-02,
3.7759e-01, 1.4439e-01, 8.5736e-02, 1.6832e-01, -4.1062e-03,
2.0312e-02, -2.0508e-01, -4.9455e-02, -1.2807e-01, -6.6944e-03,
-5.0324e-02, -4.7086e-08, -4.9077e-02, 9.1173e-02], device='cuda:0')), ('layer1.0.bn2.running_mean', tensor([ 5.0835e-08, 6.2709e-04, 9.4916e-02, -7.3116e-02, -1.6102e-02,
-3.3046e-02, -1.2526e-02, 8.5762e-02, 1.2656e-01, 7.9416e-02,
-5.5051e-02, 5.8542e-02, -2.9628e-02, 1.5995e-01, 5.1932e-02,
1.0736e-01, -6.8073e-03, 7.3404e-02, -2.1151e-01, 1.1657e-06,
1.9021e-01, 2.0322e-02, 1.2150e-01, 2.9887e-01, 1.5286e-01,
1.5446e-01, 4.2508e-01, 1.6658e-01, -1.9535e-01, 2.1023e-01,
-2.8721e-02, 8.5625e-02, 8.4987e-02, -6.4492e-02, 2.3927e-02,
-8.6530e-02, 1.4804e-01, 3.8853e-02, 1.1965e-01, 2.5974e-01,
1.2203e-01, 1.6905e-01, 2.0381e-01, -8.1762e-02, 2.4750e-01,
1.4262e-01, 5.9875e-02, -4.2531e-01, 4.5785e-02, -1.1752e-01,
-3.9317e-02, 2.4311e-02, 2.9670e-01, -1.1517e-01, -2.6087e-01,
-1.2510e-01, -6.6111e-01, 8.0603e-04, -3.2513e-02, -6.1384e-03,
-2.3477e-01, -2.5680e-08, 2.9800e-01, 2.1670e-01], device='cuda:0')), ('layer1.0.bn2.running_var', tensor([ 5.9461e-16, 1.6607e-02, 1.4680e-01, 1.9752e-02, 1.0504e-02,
5.2371e-02, 2.1798e-02, 1.3195e-02, 1.4002e-02, 7.2261e-03,
3.1354e-03, 5.9791e-02, 2.8639e-02, 1.7688e-02, 5.8631e-03,
2.7385e-02, 2.6745e-02, 8.7351e-03, 9.9867e-02, 7.6393e-13,
9.9895e-02, 2.2876e-02, 1.8758e-02, 9.9991e-02, 1.4721e-02,
3.3432e-02, 1.3276e-02, 2.8969e-02, 2.4226e-02, 2.5689e-02,
2.2515e-02, 5.2390e-02, 6.6250e-02, 1.0237e-01, 1.0278e-02,
5.5751e-03, 8.5841e-02, 2.2267e-02, 2.9026e-02, 8.4313e-02,
1.0850e-01, 7.7907e-02, 3.5947e-02, 2.6149e-02, 2.5014e-02,
1.3490e-02, 1.7496e-02, 1.1978e-02, 6.0674e-02, 1.7674e-02,
5.8531e-02, 3.4209e-02, 2.3328e-02, 4.8756e-03, 1.9696e-02,
1.6532e-02, 1.2027e-02, 5.0747e-03, 8.8367e-03, 1.2981e-04,
1.8083e-02, 4.3411e-16, 1.8637e-02, 8.6122e-02], device='cuda:0')), ('layer1.0.conv3.weight', tensor([[[[-4.0119e-09]],
[[ 1.7726e-02]],
[[ 4.9594e-02]],
...,
[[ 1.0813e-08]],
[[-1.7592e-02]],
[[-4.8830e-02]]],
[[[ 5.7415e-09]],
[[ 2.7212e-03]],
[[ 2.0600e-03]],
...,
[[ 2.3281e-09]],
[[ 3.7770e-03]],
[[-3.4820e-04]]],
[[[-2.2345e-09]],
[[-1.2277e-05]],
[[-8.5743e-03]],
...,
[[-3.2814e-09]],
[[ 7.8754e-04]],
[[-1.1076e-02]]],
...,
[[[ 6.1314e-09]],
[[-7.2828e-04]],
[[ 3.0347e-03]],
...,
[[-1.2589e-08]],
[[ 4.5773e-03]],
[[ 6.7829e-05]]],
[[[-7.4988e-09]],
[[-6.2381e-03]],
[[ 9.3970e-02]],
...,
[[ 3.6372e-09]],
[[-4.9588e-02]],
[[ 1.0847e-03]]],
[[[ 1.4321e-09]],
[[-4.5942e-03]],
[[-1.7466e-02]],
...,
[[ 6.1837e-09]],
[[ 2.9926e-02]],
[[-1.0834e-01]]]], device='cuda:0')), ('layer1.0.bn3.weight', tensor([ 6.1257e-02, 3.2458e-04, 3.6922e-02, 3.1272e-01, 5.7752e-02,
5.3033e-02, 1.1429e-01, 3.8885e-01, 9.7341e-02, 1.8650e-01,
2.7582e-01, 1.8488e-01, 2.1568e-01, 2.1534e-01, 2.3184e-01,
1.6155e-01, 2.2608e-01, 3.4452e-01, 7.7296e-02, 1.5048e-01,
1.3025e-01, 2.1159e-01, 1.1078e-06, 9.5138e-09, 1.7439e-01,
4.9364e-04, 7.4613e-03, 2.7059e-01, 2.1378e-01, 2.0946e-01,
1.7827e-01, 3.3235e-01, 2.4776e-01, 1.8123e-01, 1.2957e-01,
2.0325e-01, 1.4420e-01, 2.1640e-01, 7.3485e-04, 2.2436e-02,
-1.1337e-02, 2.1912e-01, 1.7915e-01, 2.3578e-01, 1.2075e-01,
1.5954e-05, 8.8865e-02, 1.4537e-01, 8.8577e-02, -1.0227e-02,
2.1639e-01, 3.4835e-02, 1.6429e-07, 1.3006e-01, 7.1514e-02,
1.6472e-01, 7.2320e-03, 2.2837e-01, 1.8653e-01, 2.0731e-01,
1.6906e-01, 1.5374e-01, 1.6835e-01, 9.4061e-02, 7.8185e-08,
3.9745e-03, 8.6304e-02, 2.6490e-01, 1.6982e-01, 5.5553e-03,
2.8270e-02, -2.6206e-07, 2.6815e-01, 5.8706e-02, 3.1914e-01,
5.1964e-02, 9.1686e-04, 5.5850e-02, 2.8677e-01, 1.7436e-01,
2.1439e-01, 2.4483e-01, 1.6849e-01, -4.5969e-06, 2.4238e-01,
1.7576e-01, 1.4963e-01, 7.5519e-07, 1.7493e-01, 9.7360e-07,
-7.9266e-04, 5.9956e-02, 3.1109e-03, 3.9409e-04, 2.0501e-02,
8.4718e-02, 2.0573e-01, 2.8278e-01, 9.7521e-02, 1.1674e-01,
2.6729e-01, 3.7739e-07, 2.7196e-01, 1.0523e-01, 2.5316e-01,
1.0540e-01, 2.1283e-01, 7.0997e-07, 1.5051e-02, 2.5951e-01,
8.4987e-02, 2.1008e-01, -4.4472e-03, -9.5576e-02, 2.4416e-01,
2.6090e-01, 2.8566e-01, 1.6807e-08, 2.2492e-01, 2.8301e-01,
1.8392e-01, 2.0118e-01, 2.8446e-01, 1.7130e-05, 7.6160e-02,
2.1699e-01, 1.9139e-01, 1.6825e-05, 6.9757e-02, 2.0047e-01,
1.9512e-01, 1.4369e-01, 1.6181e-01, 1.0848e-06, 1.0534e-06,
1.7272e-02, 1.3205e-01, 5.6653e-02, 2.2185e-01, 9.6189e-02,
6.2859e-02, 3.1226e-01, 7.7507e-02, 1.2070e-01, 5.8322e-02,
2.3725e-01, 2.5693e-01, 1.1205e-01, 6.6947e-02, 1.5241e-01,
8.6475e-02, 6.2732e-02, 8.2831e-02, -8.1404e-02, 1.3252e-01,
1.3642e-01, 1.3644e-01, 1.4410e-01, 2.1566e-01, -1.1276e-01,
2.8843e-01, 2.2682e-01, 2.1664e-01, -1.8349e-02, 3.6098e-02,
1.3628e-02, 3.6618e-02, 2.6597e-01, 1.5605e-01, 1.1028e-01,
-1.6914e-02, 2.4349e-01, 1.2086e-01, 1.3346e-01, 1.1930e-05,
6.3407e-02, 2.4949e-01, 2.0691e-01, 1.6867e-01, 1.5140e-01,
1.6620e-01, 6.7047e-02, 1.4889e-01, 2.4088e-01, -6.1604e-03,
-1.7191e-06, 1.6970e-01, 2.7898e-01, 2.5087e-01, 7.2350e-06,
1.1479e-01, 3.0211e-03, 8.4946e-02, 1.8868e-01, 5.3232e-03,
9.9272e-02, 1.4196e-01, 2.1632e-01, 1.9907e-01, 1.5927e-05,
2.6186e-01, 3.2927e-01, 9.0553e-07, 2.8694e-01, -4.3297e-08,
3.2081e-03, 1.4184e-01, -9.3964e-02, 1.3650e-01, 2.2932e-01,
1.5370e-01, -1.0009e-01, 9.0725e-02, 1.5219e-01, 8.0811e-07,
2.5773e-01, 1.8997e-01, 1.8119e-01, 2.0239e-01, 2.5482e-01,
-2.8486e-04, 1.5002e-01, 7.8557e-02, 5.6243e-02, 6.7529e-02,
1.7906e-01, -8.9718e-04, 1.5554e-01, 4.1687e-03, 1.4187e-02,
1.5717e-01, 2.7333e-01, 3.5062e-02, 2.7199e-01, 2.0747e-01,
6.5914e-02, 1.5763e-01, -1.5537e-03, 6.2626e-02, 4.1613e-02,
1.0547e-01, -2.1628e-02, -4.1070e-03, 2.8397e-01, 1.7258e-01,
6.1170e-02, -1.5988e-05, 7.2228e-02, 1.1727e-01, 1.6374e-01,
2.1004e-01, 5.8066e-02, 2.6768e-01, 1.4760e-03, 2.9346e-01,
1.4359e-01], device='cuda:0')), ('layer1.0.bn3.bias', tensor([-4.7521e-03, 7.3382e-02, 2.9881e-02, 5.4250e-02, -4.2102e-02,
4.5748e-02, 7.0444e-02, 1.3612e-01, 9.5970e-02, 8.6802e-02,
4.6588e-02, 8.7879e-03, -5.5630e-03, 3.5592e-02, 6.0245e-02,
2.9497e-02, 1.1487e-01, 1.0022e-01, -7.0565e-02, 5.7853e-02,
5.9134e-02, 3.1986e-02, -2.0646e-06, -1.1174e-07, 9.5474e-02,
-1.2143e-03, -1.0645e-01, -7.6647e-03, 4.3269e-02, 1.0157e-02,
3.3298e-02, 1.0682e-02, 5.0505e-02, -2.5173e-02, 5.8002e-02,
-3.9510e-02, 2.4334e-02, 2.4601e-02, 1.9136e-02, 8.7431e-02,
-1.0810e-02, 4.6122e-02, -2.8169e-03, 6.6885e-02, 3.9518e-02,
-4.4509e-05, 4.4938e-02, 1.7748e-02, -1.1751e-01, 3.7823e-02,
1.1494e-01, 1.1707e-01, -3.6071e-07, 3.7226e-02, -4.0908e-02,
8.1562e-02, 1.4487e-01, 5.9809e-03, 2.4173e-02, -1.3589e-02,
-5.1669e-02, 5.4798e-02, 4.6337e-02, 2.2271e-02, -1.4790e-07,
4.1644e-02, 3.8625e-03, 2.1603e-02, -8.6729e-03, 9.4292e-02,
-4.4369e-02, -6.0294e-07, 2.1652e-01, 8.9551e-02, 1.8707e-02,
2.0667e-02, -2.9589e-03, -2.1687e-01, 5.0260e-02, 3.8473e-02,
5.9135e-02, 3.6103e-02, 9.9497e-02, -2.0677e-05, 1.4253e-01,
1.9072e-01, 3.8042e-02, -1.2605e-06, -5.3548e-03, -2.0082e-06,
9.8597e-02, -9.4223e-02, -4.1343e-02, -8.2671e-03, -2.8833e-02,
1.0674e-01, 1.5557e-01, 2.7903e-02, 5.3139e-02, -3.1206e-02,
3.9483e-02, -7.7537e-07, 3.0037e-03, 3.5518e-02, 7.9860e-02,
2.3793e-02, -1.2981e-01, -1.3511e-06, -6.7246e-02, 1.7338e-01,
9.5783e-02, 9.1839e-02, -4.3202e-02, 7.8402e-02, -6.3475e-05,
5.4036e-02, 4.6969e-02, -2.4410e-07, 1.5709e-01, 2.9272e-02,
3.0758e-02, -1.5072e-01, 9.3047e-02, -7.6240e-05, 1.1170e-01,
1.5533e-02, 6.0914e-02, -5.9870e-04, 2.6366e-02, 5.2601e-02,
1.6601e-01, -3.9785e-02, 5.7633e-02, -2.7045e-06, -3.3708e-06,
1.1546e-01, 4.9188e-02, -1.4883e-02, 4.4428e-02, -3.1150e-02,
-2.3653e-02, 6.5779e-02, 6.7056e-02, 9.8390e-02, -2.3785e-01,
3.3035e-02, 1.9017e-01, 1.6324e-02, -3.0738e-01, 3.1187e-02,
-2.1642e-01, 1.1853e-02, -4.2425e-03, 5.6296e-03, -7.5315e-02,
1.3189e-02, 2.1798e-02, 6.3251e-04, 4.8708e-03, -5.7906e-02,
5.1894e-02, 1.3344e-01, -1.5952e-02, 1.6532e-01, -2.8305e-02,
4.0512e-03, 2.7601e-02, 1.2813e-02, -1.5498e-03, 2.9502e-02,
8.6315e-02, -4.0821e-02, -3.2629e-02, -1.5396e-02, -1.8829e-05,
1.0559e-02, 3.5416e-02, 1.2415e-01, 7.7687e-02, -1.1105e-02,
1.8961e-02, -5.8442e-02, -1.7247e-02, -4.9154e-02, 1.1553e-02,
-2.0810e-06, 1.3872e-02, 1.0846e-01, 6.8531e-02, 1.7672e-01,
3.5634e-02, -4.0294e-02, 4.6163e-02, 3.3498e-02, 2.2283e-02,
4.6689e-02, 2.9249e-02, 1.0380e-01, 2.7987e-02, -4.7313e-05,
8.3607e-02, 7.9628e-02, -2.0540e-06, 1.2765e-01, -2.4337e-07,
1.1281e-01, -1.0358e-02, -4.9932e-03, 9.1279e-02, 5.0166e-02,
-1.9497e-02, -3.3888e-04, -3.0161e-02, 7.7116e-02, -2.9716e-06,
7.0705e-02, 2.3647e-02, 3.5755e-02, -6.3678e-02, 1.0491e-01,
1.1383e-01, 1.3069e-03, -4.9135e-02, 1.1538e-01, -6.0227e-02,
9.9392e-04, 6.4781e-02, 7.6379e-02, 5.9853e-02, 1.1961e-01,
-5.1671e-02, 5.2803e-02, 2.7144e-02, 5.1999e-02, 1.9074e-01,
1.7143e-02, 5.5178e-02, 1.2572e-01, -2.6754e-03, -3.1276e-02,
4.7069e-02, 5.1690e-02, -5.9475e-03, 2.9633e-02, 1.9122e-01,
1.4001e-02, -4.0885e-04, 2.3924e-02, -3.6612e-02, 3.5178e-02,
8.2615e-02, 6.9214e-02, 1.1781e-01, 3.4332e-02, 1.9473e-01,
3.2748e-02], device='cuda:0')), ('layer1.0.bn3.running_mean', tensor([ 2.0184e-02, 4.5334e-04, -8.9353e-03, -7.5252e-02, -6.1524e-03,
1.4150e-02, -6.0100e-02, 4.9168e-02, 2.4714e-02, -5.3570e-02,
-4.5486e-02, 2.2949e-02, 2.7561e-02, 7.9998e-02, 2.7238e-02,
9.2494e-03, -1.1641e-01, 3.8745e-02, 5.0605e-03, -1.3659e-02,
-7.2709e-02, 1.1049e-01, 1.3841e-06, -1.1598e-07, -1.2370e-02,
-1.0110e-04, 1.5542e-02, 3.3336e-02, -1.9242e-01, 7.9447e-02,
-9.6315e-02, -4.2818e-03, -4.5212e-02, -5.0333e-02, -7.4657e-04,
-1.0212e-02, -1.7877e-02, -3.2092e-03, -1.5007e-02, 3.1666e-03,
3.7117e-03, -2.1898e-02, -1.1164e-01, 1.7101e-02, -2.5701e-02,
1.7586e-05, 5.6353e-03, -8.0848e-04, -1.3683e-02, 7.4592e-04,
-1.6695e-03, -1.7781e-02, -2.7847e-07, 2.8052e-02, 6.5875e-02,
5.7626e-02, -7.5375e-03, -9.4141e-02, 5.3709e-02, 6.6873e-02,
2.7402e-02, -5.7566e-02, -7.8638e-02, -2.1565e-02, -5.8122e-08,
-1.4108e-03, 3.3652e-02, -6.0796e-02, 9.2371e-02, -6.6974e-03,
-2.7073e-02, -1.4258e-07, -6.1510e-02, 8.1816e-02, 7.2517e-02,
3.6931e-02, 2.1227e-03, 6.5289e-03, 9.1030e-02, 6.3455e-02,
4.1044e-02, 1.8491e-01, -1.0873e-01, -1.0806e-05, 3.2235e-02,
-5.7058e-02, -1.5838e-02, 7.6316e-07, -3.5413e-02, 8.9018e-07,
9.4534e-03, 2.6818e-02, -6.9917e-03, -1.2908e-04, -1.1645e-03,
-1.6912e-02, -1.2214e-01, 8.1981e-02, 1.2614e-02, -1.9792e-04,
5.7958e-02, -7.2069e-07, -1.5860e-02, -1.0802e-02, -4.5527e-02,
2.4828e-02, -1.4175e-02, 3.6009e-07, 1.1796e-02, 2.1772e-02,
5.0924e-02, 2.8583e-02, -1.4605e-02, 1.4115e-02, 2.4587e-02,
-2.5819e-02, 5.1181e-02, 2.5541e-08, -7.7069e-04, -1.0337e-02,
-7.9262e-02, -1.6196e-02, -4.8100e-02, 5.1066e-05, 2.4413e-02,
2.7514e-02, 5.4527e-02, -9.1486e-05, -1.0971e-01, -3.5396e-03,
-5.2528e-02, 7.5389e-02, -3.5125e-02, 5.3588e-08, 8.3788e-07,
5.0161e-03, -4.4067e-03, -4.0199e-02, 6.1700e-02, 3.2923e-02,
-3.2054e-02, 3.8194e-02, -4.7097e-02, -4.5798e-02, -7.9202e-02,
-3.7634e-02, -6.3261e-02, -2.2658e-02, -2.4417e-02, 1.3351e-02,
8.0719e-02, 3.0532e-02, 3.3867e-02, 5.7184e-03, 3.9517e-02,
7.7062e-03, 7.3452e-02, -5.3931e-02, -9.5099e-02, -5.9494e-02,
2.3852e-01, -3.4742e-02, -2.1182e-02, 8.6606e-05, 6.1238e-03,
-2.8897e-02, -1.1929e-02, 2.0336e-01, -3.6211e-02, 3.0506e-02,
-3.3776e-03, -6.7834e-02, 1.2732e-02, 5.5975e-02, -1.0137e-06,
-2.6809e-02, -3.5270e-02, 1.0178e-01, 4.5295e-02, 3.0662e-03,
7.2206e-03, -2.6278e-02, 1.7373e-02, -1.6618e-02, -1.2031e-02,
7.1543e-07, 2.9507e-02, 2.3738e-02, -1.7306e-02, 8.4228e-03,
-6.9493e-03, -6.9199e-03, 1.8662e-03, -7.1136e-02, -2.3724e-03,
-2.7569e-02, 2.7457e-03, -1.8368e-01, -1.3498e-01, 1.8480e-05,
7.7467e-03, 5.2126e-02, 2.2500e-07, 1.8811e-01, -3.9934e-09,
-6.3393e-03, 5.7957e-02, 1.9258e-02, -2.2388e-01, 2.9520e-02,
6.3161e-02, 3.4584e-02, 2.3850e-02, -7.5089e-02, -2.3928e-06,
1.1528e-01, -5.1541e-02, 7.0340e-02, 4.1758e-03, -6.4713e-03,
-7.4174e-04, -4.5243e-02, 4.0725e-03, 2.9000e-02, 7.8960e-02,
3.7333e-02, -8.0464e-03, -1.2227e-02, -7.5422e-03, -9.4457e-03,
-1.6058e-02, 2.2890e-01, 4.7060e-02, 2.0861e-02, 2.1091e-01,
-1.0997e-02, 2.8849e-03, 1.0043e-02, 2.3445e-02, 3.2478e-02,
-7.1065e-02, 2.5317e-02, -1.5693e-03, 3.4646e-02, -2.6326e-02,
-2.6507e-02, -4.4122e-05, -8.7328e-03, 4.0251e-02, 7.8662e-03,
-5.4224e-02, 4.0470e-02, -1.2639e-01, 1.4116e-02, -2.6492e-02,
-3.5285e-02], device='cuda:0')), ('layer1.0.bn3.running_var', tensor([ 5.9753e-03, 2.9410e-04, 2.2775e-04, 2.5453e-02, 2.4658e-04,
2.1143e-03, 3.8699e-03, 3.7719e-02, 1.1424e-02, 7.5526e-03,
9.3138e-03, 7.3118e-03, 9.7306e-03, 2.4725e-02, 1.3871e-02,
5.7549e-03, 9.5564e-03, 2.2274e-02, 5.5561e-04, 1.9365e-02,
3.3895e-03, 4.5192e-03, 4.0072e-13, 1.5298e-14, 1.3110e-02,
2.5759e-07, 2.2922e-04, 7.2376e-03, 7.1590e-03, 5.6691e-03,
4.5992e-03, 1.7805e-02, 1.7653e-02, 3.1532e-03, 4.5118e-03,
1.1456e-02, 8.9262e-03, 8.3100e-03, 5.7004e-04, 2.2457e-03,
1.7300e-04, 1.5967e-02, 1.1595e-02, 1.6007e-02, 6.7701e-03,
9.1658e-11, 3.2120e-03, 3.1561e-03, 5.7093e-04, 1.6867e-04,
1.0356e-02, 1.0258e-03, 4.1103e-14, 3.2948e-03, 1.7775e-03,
1.4259e-02, 2.2513e-04, 9.7883e-03, 8.9819e-03, 8.0331e-03,
3.5536e-03, 4.0108e-03, 7.0689e-03, 3.2281e-03, 2.2499e-14,
3.1522e-04, 2.4139e-03, 5.9923e-03, 3.4887e-03, 2.0722e-04,
2.8181e-04, 9.5322e-14, 1.2933e-02, 1.1607e-02, 2.2426e-02,
1.3298e-03, 2.5265e-06, 9.4097e-04, 1.0331e-02, 5.0758e-03,
1.5776e-02, 1.3721e-02, 1.1432e-02, 3.7877e-11, 9.9955e-03,
1.5599e-02, 5.7056e-03, 1.9402e-13, 1.5418e-02, 1.7986e-13,
6.0215e-04, 9.5347e-04, 1.8957e-04, 3.4289e-06, 1.6054e-04,
1.9464e-03, 1.3862e-02, 8.2601e-03, 3.1031e-03, 3.5196e-03,
1.5433e-02, 1.1903e-13, 1.4601e-02, 4.7773e-03, 1.9037e-02,
5.8789e-03, 2.5534e-03, 6.0355e-13, 2.0346e-04, 1.6413e-02,
1.2815e-02, 6.6989e-03, 1.5945e-04, 4.9616e-03, 8.1303e-03,
8.5662e-03, 1.5269e-02, 1.2781e-14, 1.6640e-02, 2.1393e-02,
5.1683e-03, 2.9232e-03, 1.7922e-02, 1.0104e-08, 5.8175e-03,
7.1051e-03, 5.7812e-03, 5.9970e-08, 9.1011e-04, 6.2531e-03,
1.6075e-02, 1.9789e-03, 7.1269e-03, 1.4261e-12, 5.7846e-12,
3.6986e-04, 2.5142e-03, 1.1083e-03, 1.3133e-02, 5.1764e-04,
1.0059e-03, 9.6899e-03, 3.6846e-03, 4.2558e-03, 1.7454e-03,
1.3268e-02, 1.6535e-02, 5.4452e-03, 9.1705e-04, 3.3724e-03,
2.5224e-03, 1.0499e-03, 1.3601e-03, 1.3572e-03, 2.5463e-03,
3.3373e-03, 3.5869e-03, 3.6988e-03, 6.2769e-03, 1.9938e-03,
8.4959e-03, 2.2239e-02, 6.7073e-03, 9.0879e-04, 2.4540e-04,
1.5888e-04, 1.0354e-03, 1.5766e-02, 6.2222e-03, 5.0267e-03,
5.8083e-04, 5.1943e-03, 1.3083e-03, 2.6041e-03, 1.3887e-10,
1.2750e-03, 1.0255e-02, 2.0926e-02, 4.4365e-03, 5.7806e-03,
5.5205e-03, 1.4726e-03, 7.9955e-03, 5.3122e-03, 2.5704e-04,
1.4809e-12, 6.6627e-03, 2.5483e-02, 1.0330e-02, 4.3661e-04,
1.1381e-02, 1.7797e-04, 1.6772e-02, 6.9293e-03, 1.9882e-04,
7.5829e-03, 9.5277e-03, 8.8398e-03, 4.5115e-03, 8.0091e-10,
2.1312e-02, 1.8472e-02, 4.5033e-13, 2.1182e-02, 1.3137e-14,
1.9543e-04, 3.1692e-03, 1.2520e-03, 4.4196e-03, 8.7075e-03,
7.3718e-03, 2.6734e-03, 1.3873e-03, 4.8220e-03, 3.3179e-12,
2.2015e-02, 3.9697e-03, 6.4724e-03, 7.7465e-03, 1.3029e-02,
1.0232e-03, 2.4010e-03, 6.2523e-04, 1.2158e-02, 7.7891e-04,
7.2560e-03, 3.1088e-04, 5.2979e-03, 2.7234e-04, 6.3333e-04,
8.9140e-04, 1.4879e-02, 5.7939e-04, 2.2887e-02, 6.1351e-03,
1.0504e-03, 5.0027e-03, 1.1102e-03, 8.0596e-04, 5.7298e-04,
1.4311e-03, 4.2223e-04, 4.4093e-04, 1.6453e-02, 1.7354e-02,
2.5492e-03, 1.2703e-08, 7.0726e-04, 2.0684e-03, 4.1034e-03,
1.4209e-02, 1.1267e-02, 1.8433e-02, 2.0710e-04, 3.0578e-02,
2.8345e-03], device='cuda:0')), ('layer1.0.downsample.0.weight', tensor([[[[ 7.5811e-03]],
[[-1.6594e-01]],
[[ 9.1546e-03]],
...,
[[-2.2368e-02]],
[[ 4.2574e-02]],
[[ 1.4474e-03]]],
[[[ 1.2800e-02]],
[[-1.1457e-02]],
[[-2.5123e-02]],
...,
[[-4.1692e-04]],
[[ 2.2987e-02]],
[[ 2.1549e-03]]],
[[[-3.5923e-02]],
[[ 1.3393e-02]],
[[-2.6259e-02]],
...,
[[ 5.9477e-02]],
[[-2.7209e-04]],
[[-6.3618e-03]]],
...,
[[[-3.9645e-02]],
[[-3.9428e-01]],
[[ 7.2763e-02]],
...,
[[-1.0293e-02]],
[[ 1.1058e-01]],
[[ 1.6492e-02]]],
[[[ 1.5314e-03]],
[[ 1.1643e-02]],
[[ 6.7886e-02]],
...,
[[-7.0245e-02]],
[[ 1.6036e-01]],
[[ 3.6155e-02]]],
[[[ 2.0393e-02]],
[[ 3.6532e-02]],
[[-3.5239e-02]],
...,
[[-1.1597e-01]],
[[ 2.4156e-02]],
[[-5.0600e-02]]]], device='cuda:0')), ('layer1.0.downsample.1.weight', tensor([ 2.4457e-01, 2.2984e-01, 1.4612e-01, 3.4260e-01, 2.5927e-02,
1.1437e-01, 2.1102e-01, 4.5606e-01, 1.8027e-01, 1.5431e-01,
3.5582e-01, 2.1846e-01, 2.5678e-01, 1.5123e-02, 4.0578e-01,
2.0874e-01, 9.8095e-02, 3.9442e-01, 1.9774e-01, 3.0000e-01,
2.4856e-01, 2.3178e-01, -1.0182e-08, 1.3827e-08, 3.1164e-01,
4.8934e-04, 2.6274e-01, 5.9192e-02, 1.9006e-01, 2.0053e-01,
1.6345e-01, 3.0634e-01, 3.5806e-01, 1.8923e-02, 3.0784e-01,
4.3668e-01, 2.7498e-01, 1.6151e-01, 2.9890e-01, 1.2460e-01,
7.0893e-02, 3.3482e-01, 3.4934e-01, 2.8829e-01, 2.7963e-01,
8.1415e-07, 3.6592e-01, 1.6732e-01, 2.2241e-01, 1.9294e-01,
2.9332e-01, 1.4812e-01, 2.0370e-07, 1.6793e-01, 2.2183e-01,
2.7442e-01, -1.1456e-02, 3.9355e-01, 2.5354e-01, 3.2749e-01,
3.3727e-01, 2.0058e-01, 2.4418e-01, 1.6629e-01, 2.0949e-08,
1.8675e-01, 1.8726e-01, 2.4386e-01, 1.7288e-01, 2.2883e-01,
1.9265e-01, 1.8148e-07, 1.8998e-01, 1.9738e-01, 3.0906e-01,
1.1432e-01, 2.7538e-03, 2.8674e-01, 2.9543e-01, 2.0380e-01,
2.9774e-01, 3.0047e-01, 3.0071e-01, 3.0339e-06, 2.5298e-01,
2.8255e-01, 2.3041e-01, -2.2464e-07, 9.0851e-02, 9.4888e-08,
2.3342e-01, 2.8756e-01, 2.5694e-01, 1.9065e-03, 1.9803e-02,
1.2581e-01, 2.5297e-01, 1.0138e-01, 1.9470e-01, 2.4283e-01,
2.2812e-01, 3.9112e-07, 4.3694e-01, 2.0982e-01, 3.3158e-01,
3.6135e-01, 1.6712e-01, 2.8431e-07, 2.4639e-01, 3.1374e-01,
2.0042e-01, 1.7867e-01, 2.4927e-01, 2.6893e-01, 2.4301e-02,
3.3343e-01, 2.8018e-01, 1.0910e-07, 3.1720e-01, 3.2437e-01,
1.7580e-01, 2.1966e-01, 2.2206e-01, 3.8821e-05, 2.1403e-01,
3.1898e-01, 2.2020e-01, 2.1191e-04, 2.4318e-01, 3.2712e-01,
3.5392e-01, 1.9045e-01, 2.4259e-01, 5.9542e-07, 2.1926e-06,
2.0136e-01, 2.0292e-01, 1.6009e-01, 3.1163e-01, 3.1269e-02,
1.5136e-01, 3.4167e-01, 2.2320e-01, 1.7928e-01, 3.0206e-01,
3.3896e-01, 3.2244e-01, 3.3214e-01, 3.1712e-01, 2.6545e-01,
3.4695e-01, 1.9813e-01, 1.5368e-01, 2.3095e-01, 1.2709e-01,
2.2131e-01, 7.9861e-02, 1.0763e-01, 3.4360e-01, 2.8470e-01,
3.1999e-01, 3.4235e-01, 2.8543e-01, 2.5500e-01, 8.0065e-02,
2.1822e-01, 1.2625e-01, 3.3813e-01, 3.6787e-01, 2.0902e-01,
7.5418e-02, 3.0735e-01, 4.3765e-02, 9.5508e-02, -3.7830e-06,
3.1341e-01, 2.3747e-01, 2.9429e-01, 1.7653e-01, 3.3874e-01,
2.2616e-01, 1.9847e-01, 3.5470e-01, 8.1458e-03, 2.0669e-01,
1.0625e-06, 3.6871e-01, 3.6358e-01, 1.9882e-01, 1.6871e-01,
3.9717e-01, 2.4319e-01, 2.2733e-01, 2.5196e-01, 4.8692e-02,
3.4219e-01, 3.5018e-01, -4.2149e-02, 1.8716e-01, -7.9278e-06,
2.8354e-01, 3.7017e-01, 1.4341e-07, 3.1774e-01, 7.5645e-08,
1.6309e-01, 1.6707e-01, 2.2129e-01, 2.7987e-01, 2.5573e-01,
2.8884e-01, 2.4309e-01, 8.3486e-02, 2.5541e-01, 6.3526e-07,
3.6438e-01, 2.3440e-01, 2.2243e-01, 3.1752e-01, 3.1876e-01,
2.5398e-01, 1.6543e-01, 6.8547e-02, 1.9213e-01, 1.8097e-01,
2.8546e-01, 2.0063e-01, 1.9390e-01, 2.0524e-01, 1.8420e-01,
2.0247e-02, 2.7564e-01, 1.4836e-01, 3.3919e-01, 1.8534e-01,
2.9466e-01, 2.7174e-01, 2.9968e-01, 1.3451e-01, 1.9264e-01,
1.2815e-01, 1.7942e-01, 3.1054e-01, 3.0394e-01, 3.0231e-01,
2.6601e-01, 7.0521e-05, 1.6745e-01, 2.4501e-01, 1.2502e-01,
3.6441e-01, 2.4699e-01, 3.7460e-01, 1.6707e-01, 3.6041e-01,
1.8811e-01], device='cuda:0')), ('layer1.0.downsample.1.bias', tensor([-4.7521e-03, 7.3382e-02, 2.9881e-02, 5.4250e-02, -4.2102e-02,
4.5748e-02, 7.0444e-02, 1.3612e-01, 9.5970e-02, 8.6802e-02,
4.6588e-02, 8.7879e-03, -5.5630e-03, 3.5592e-02, 6.0245e-02,
2.9497e-02, 1.1487e-01, 1.0022e-01, -7.0565e-02, 5.7853e-02,
5.9134e-02, 3.1986e-02, -2.0646e-06, -1.1174e-07, 9.5474e-02,
-1.2143e-03, -1.0645e-01, -7.6647e-03, 4.3269e-02, 1.0157e-02,
3.3298e-02, 1.0682e-02, 5.0505e-02, -2.5173e-02, 5.8002e-02,
-3.9510e-02, 2.4334e-02, 2.4601e-02, 1.9136e-02, 8.7431e-02,
-1.0810e-02, 4.6122e-02, -2.8169e-03, 6.6885e-02, 3.9518e-02,
-4.4509e-05, 4.4938e-02, 1.7748e-02, -1.1751e-01, 3.7823e-02,
1.1494e-01, 1.1707e-01, -3.6071e-07, 3.7226e-02, -4.0908e-02,
8.1562e-02, 1.4487e-01, 5.9809e-03, 2.4173e-02, -1.3589e-02,
-5.1669e-02, 5.4798e-02, 4.6337e-02, 2.2271e-02, -1.4790e-07,
4.1644e-02, 3.8625e-03, 2.1603e-02, -8.6729e-03, 9.4292e-02,
-4.4369e-02, -6.0294e-07, 2.1652e-01, 8.9551e-02, 1.8707e-02,
2.0667e-02, -2.9589e-03, -2.1687e-01, 5.0260e-02, 3.8473e-02,
5.9135e-02, 3.6103e-02, 9.9497e-02, -2.0677e-05, 1.4253e-01,
1.9072e-01, 3.8042e-02, -1.2605e-06, -5.3548e-03, -2.0082e-06,
9.8597e-02, -9.4223e-02, -4.1343e-02, -8.2671e-03, -2.8833e-02,
1.0674e-01, 1.5557e-01, 2.7903e-02, 5.3139e-02, -3.1206e-02,
3.9483e-02, -7.7537e-07, 3.0037e-03, 3.5518e-02, 7.9860e-02,
2.3793e-02, -1.2981e-01, -1.3511e-06, -6.7246e-02, 1.7338e-01,
9.5783e-02, 9.1839e-02, -4.3202e-02, 7.8402e-02, -6.3475e-05,
5.4036e-02, 4.6969e-02, -2.4410e-07, 1.5709e-01, 2.9272e-02,
3.0758e-02, -1.5072e-01, 9.3047e-02, -7.6240e-05, 1.1170e-01,
1.5533e-02, 6.0914e-02, -5.9870e-04, 2.6366e-02, 5.2601e-02,
1.6601e-01, -3.9785e-02, 5.7633e-02, -2.7045e-06, -3.3708e-06,
1.1546e-01, 4.9188e-02, -1.4883e-02, 4.4428e-02, -3.1150e-02,
-2.3653e-02, 6.5779e-02, 6.7056e-02, 9.8390e-02, -2.3785e-01,
3.3035e-02, 1.9017e-01, 1.6324e-02, -3.0738e-01, 3.1187e-02,
-2.1642e-01, 1.1853e-02, -4.2425e-03, 5.6296e-03, -7.5315e-02,
1.3189e-02, 2.1798e-02, 6.3251e-04, 4.8708e-03, -5.7906e-02,
5.1894e-02, 1.3344e-01, -1.5952e-02, 1.6532e-01, -2.8305e-02,
4.0512e-03, 2.7601e-02, 1.2813e-02, -1.5498e-03, 2.9502e-02,
8.6315e-02, -4.0821e-02, -3.2629e-02, -1.5396e-02, -1.8829e-05,
1.0559e-02, 3.5416e-02, 1.2415e-01, 7.7687e-02, -1.1105e-02,
1.8961e-02, -5.8442e-02, -1.7247e-02, -4.9154e-02, 1.1553e-02,
-2.0810e-06, 1.3872e-02, 1.0846e-01, 6.8531e-02, 1.7672e-01,
3.5634e-02, -4.0294e-02, 4.6163e-02, 3.3498e-02, 2.2283e-02,
4.6689e-02, 2.9249e-02, 1.0380e-01, 2.7987e-02, -4.7313e-05,
8.3607e-02, 7.9628e-02, -2.0540e-06, 1.2765e-01, -2.4337e-07,
1.1281e-01, -1.0358e-02, -4.9932e-03, 9.1279e-02, 5.0166e-02,
-1.9497e-02, -3.3888e-04, -3.0161e-02, 7.7116e-02, -2.9716e-06,
7.0705e-02, 2.3647e-02, 3.5755e-02, -6.3678e-02, 1.0491e-01,
1.1383e-01, 1.3069e-03, -4.9135e-02, 1.1538e-01, -6.0227e-02,
9.9392e-04, 6.4781e-02, 7.6379e-02, 5.9853e-02, 1.1961e-01,
-5.1671e-02, 5.2803e-02, 2.7144e-02, 5.1999e-02, 1.9074e-01,
1.7143e-02, 5.5178e-02, 1.2572e-01, -2.6754e-03, -3.1276e-02,
4.7069e-02, 5.1690e-02, -5.9475e-03, 2.9633e-02, 1.9122e-01,
1.4001e-02, -4.0885e-04, 2.3924e-02, -3.6612e-02, 3.5178e-02,
8.2615e-02, 6.9214e-02, 1.1781e-01, 3.4332e-02, 1.9473e-01,
3.2748e-02], device='cuda:0')), ('layer1.0.downsample.1.running_mean', tensor([-3.3582e-01, -3.2734e-02, 9.2928e-02, -2.3341e-01, -1.4776e-02,
-2.4913e-01, -3.0179e-01, 1.4513e-01, -7.4187e-01, -4.0522e-01,
-4.9738e-01, -2.3839e-02, 1.0634e-02, 1.7370e-01, -2.0598e-01,
-1.8817e-01, -2.9287e-01, 1.5948e-01, 3.4533e-01, -7.9472e-01,
-3.6631e-01, 5.5492e-02, 1.4413e-06, 6.3116e-07, -8.9981e-02,
-1.0940e-03, 2.8230e-01, -1.9937e-01, 4.5423e-02, 1.1463e-01,
2.2490e-02, -6.8430e-02, -1.0595e-01, -1.2882e-02, -4.5302e-01,
-5.9576e-01, -6.5497e-01, 1.4108e-01, -5.5488e-01, -1.7652e-01,
4.5116e-02, -2.8137e-01, -2.3714e-01, -8.5441e-02, -2.2772e-01,
3.4473e-05, -4.0135e-01, -2.1367e-01, -3.9226e-02, -5.8393e-01,
-3.9183e-01, 3.6220e-02, -1.4943e-06, 1.4087e-02, -2.6330e-01,
-8.6968e-01, -1.1263e-01, -4.1946e-02, 9.7181e-02, -2.6787e-01,
-6.3795e-01, -8.2990e-02, -2.0244e-01, -3.0591e-01, -8.3373e-07,
-7.1344e-02, -2.4399e-01, 4.2774e-03, 8.6159e-03, -3.7885e-01,
-2.4281e-01, -2.7559e-07, -8.1162e-01, -3.8792e-01, -1.8653e-01,
3.8799e-01, -8.4840e-03, 2.2213e-01, -3.9215e-01, 1.9764e-01,
-4.9439e-01, 1.5703e-01, -4.0676e-01, 6.3938e-05, 8.2878e-01,
1.8170e-01, -3.2141e-01, -6.2356e-07, 1.5982e-01, -7.6148e-07,
-4.2822e-01, -2.0748e-01, 6.4092e-03, 1.7699e-02, 3.9913e-03,
-4.7021e-01, 2.6191e-01, 3.8090e-01, -7.7878e-02, -3.9128e-01,
-2.3662e-01, -1.4194e-06, -3.0428e-01, -5.5854e-01, -1.9086e-01,
-4.2683e-01, 5.5728e-03, -2.1425e-06, -4.3826e-02, -8.1340e-02,
-3.6905e-01, 4.0619e-01, 4.3619e-02, -1.7492e-01, -1.3909e-01,
-5.2995e-02, -3.2330e-01, -4.1767e-07, -5.2748e-02, -1.0260e-01,
-3.8758e-01, -3.2681e-02, -3.1551e-01, -2.1463e-04, -4.0721e-01,
-1.3137e-01, -3.1778e-01, 1.9749e-03, 6.2323e-02, 7.7641e-02,
-2.5466e-02, -1.7477e-01, 1.3220e-01, -2.1997e-06, -1.5769e-05,
-3.1272e-01, 1.4780e-01, -4.4415e-02, -2.6772e-01, -7.2281e-02,
-1.2964e-01, -1.2267e-01, -7.2904e-02, -4.7452e-01, -4.4261e-03,
-3.8157e-01, 3.2064e-02, -1.6148e-01, 2.8321e-02, -3.3088e-01,
-1.2691e-01, -3.3993e-02, 1.2466e-01, 7.1399e-02, 2.1701e-01,
-2.1595e-01, -1.2807e-01, -2.0995e-03, -2.4545e-01, -5.1459e-01,
-2.4458e-01, 2.4112e-01, -1.8517e-01, 1.6860e+00, 4.8322e-02,
-3.8258e-01, 2.1185e-02, -7.1561e-01, -2.9062e-01, -2.1116e-01,
-5.4396e-02, -4.0527e-01, 7.3901e-02, -8.9251e-02, -3.7257e-05,
-2.6419e-03, 2.5726e-01, 9.5989e-01, 4.7688e-01, -3.6548e-01,
1.5424e-01, 2.2409e-01, -3.4355e-01, -1.0373e-01, 4.5298e-02,
-2.2673e-06, -7.4707e-01, -2.4686e-01, -1.1831e-01, 1.7758e-01,
-4.6059e-01, 2.2237e-01, -2.0899e-01, -2.1346e-01, -8.8123e-02,
-1.0356e+00, -2.9040e-01, 3.0804e-01, -9.2490e-02, 7.0288e-05,
-3.3769e-01, 1.9198e-01, 7.7345e-07, -1.0055e-02, 7.1525e-08,
-3.4565e-01, -1.0906e-01, 3.6143e-03, -2.9875e-01, -1.3478e-01,
-2.7061e-01, 7.0553e-02, -3.4871e-02, -3.6797e-01, 2.1517e-06,
-2.1074e-01, 7.4181e-02, 1.1791e-01, -2.4863e-01, 3.6520e-01,
-1.8940e-01, 1.6151e-01, -8.9373e-02, -4.2941e-01, 1.6770e-01,
-5.4754e-01, 1.0610e-01, -4.2902e-01, 1.0134e-01, 2.4417e-01,
5.1096e-02, -6.3823e-02, -2.5318e-01, 1.0381e-01, 4.4021e-01,
-8.2747e-01, 3.9008e-01, 1.4637e-01, -3.1890e-01, -1.1701e-01,
-2.0046e-01, -6.2476e-01, -4.1308e-01, -9.3190e-02, -9.8484e-02,
-4.4976e-01, -2.0317e-04, 2.0481e-01, -3.4697e-01, 2.2672e-01,
-1.4442e-01, -3.2782e-01, -9.4906e-01, 1.5404e-01, 2.2443e-01,
1.4919e-01], device='cuda:0')), ('layer1.0.downsample.1.running_var', tensor([ 4.2576e-02, 8.2049e-02, 7.4886e-03, 1.2342e-01, 1.0331e-03,
3.6497e-02, 3.9575e-02, 4.6431e-02, 3.2680e-02, 2.6248e-02,
4.6396e-02, 2.9158e-02, 2.6550e-02, 1.1069e-02, 8.0318e-02,
2.2528e-02, 2.7441e-02, 3.7461e-02, 1.9742e-02, 4.4432e-02,
3.3360e-02, 1.9381e-02, 4.6864e-13, 4.5202e-14, 6.7372e-02,
1.6406e-06, 3.5863e-02, 4.2062e-03, 3.7249e-02, 3.6191e-02,
3.1338e-02, 5.5185e-02, 7.8041e-02, 1.0677e-03, 4.5144e-02,
6.5771e-02, 5.1010e-02, 3.2281e-02, 1.1826e-01, 3.1323e-02,
2.8810e-03, 6.8764e-02, 6.3108e-02, 7.9717e-02, 3.5944e-02,
1.8554e-10, 6.1870e-02, 8.1551e-03, 5.2128e-03, 1.5544e-02,
4.9164e-02, 2.7460e-02, 3.1095e-13, 1.6923e-02, 3.3910e-02,
2.6321e-02, 1.5237e-03, 2.6231e-02, 5.9430e-02, 5.0749e-02,
1.2441e-02, 3.2874e-02, 2.7709e-02, 2.2240e-02, 1.8722e-13,
8.0735e-02, 2.1360e-02, 3.5947e-02, 1.2723e-02, 4.9485e-02,
4.1818e-03, 1.4822e-12, 7.0186e-02, 7.8661e-02, 1.0946e-01,
8.6640e-03, 2.5458e-05, 2.1767e-02, 3.7485e-02, 3.1092e-02,
7.1144e-02, 7.1728e-02, 6.8895e-02, 5.9931e-10, 9.5977e-02,
4.9123e-02, 2.7610e-02, 6.1441e-14, 2.7523e-02, 1.2197e-13,
1.2396e-01, 7.5245e-03, 5.0878e-02, 1.4153e-04, 1.3011e-03,
8.1803e-03, 1.4248e-02, 1.5328e-02, 2.3838e-02, 3.7134e-02,
5.8370e-02, 5.6410e-13, 8.2700e-02, 1.5140e-02, 6.1327e-02,
8.6041e-02, 5.1501e-03, 7.6795e-13, 3.7475e-02, 7.4407e-02,
7.6286e-02, 4.3459e-02, 5.8441e-02, 3.9704e-02, 3.2716e-03,
4.0273e-02, 6.8190e-02, 4.8041e-14, 5.2347e-02, 9.3200e-02,
3.6727e-02, 6.0229e-03, 5.7488e-02, 1.7465e-08, 5.3080e-02,
3.5779e-02, 1.9248e-02, 8.5485e-07, 1.2829e-02, 4.6224e-02,
1.0131e-01, 1.4054e-02, 2.9316e-02, 1.1926e-12, 1.8417e-11,
3.7068e-02, 2.2129e-02, 5.7356e-03, 8.1125e-02, 1.7039e-03,
4.8325e-03, 6.5224e-02, 2.2701e-02, 2.3557e-02, 1.1503e-02,
7.1553e-02, 7.1058e-02, 3.6570e-02, 1.2836e-02, 1.5822e-02,
2.1518e-02, 5.1881e-03, 1.4487e-02, 8.7995e-03, 1.0897e-02,
2.8639e-02, 1.1333e-02, 5.2220e-03, 4.6166e-02, 1.7697e-02,
4.6317e-02, 5.9265e-02, 1.4189e-02, 2.9225e-01, 2.8139e-03,
2.8084e-02, 1.0752e-02, 1.6471e-02, 8.8671e-02, 1.9352e-02,
1.1743e-02, 3.9949e-02, 2.4881e-03, 7.8021e-03, 1.8516e-10,
1.1910e-02, 4.8778e-02, 3.1828e-02, 2.8010e-02, 7.1716e-02,
1.7439e-02, 2.0412e-02, 5.7679e-02, 1.5171e-03, 6.3709e-02,
9.7580e-12, 4.8931e-02, 4.6674e-02, 4.5978e-02, 1.0970e-01,
1.4755e-01, 5.3101e-02, 1.0809e-01, 4.0907e-02, 4.0855e-03,
1.5600e-02, 7.5194e-02, 1.0339e-02, 2.2269e-02, 1.1132e-09,
8.7459e-02, 3.1863e-02, 1.3688e-13, 5.5655e-02, 6.1422e-14,
2.0547e-02, 1.6374e-02, 8.6761e-03, 3.1495e-02, 4.2383e-02,
4.3731e-02, 1.5849e-02, 2.8498e-03, 2.5285e-02, 4.0346e-12,
5.5747e-02, 2.3284e-02, 2.7143e-02, 3.7541e-02, 3.4432e-02,
2.0901e-01, 5.3354e-03, 1.3701e-03, 9.2496e-02, 7.4451e-03,
6.3354e-02, 7.7451e-02, 2.2226e-02, 3.9656e-02, 3.4413e-02,
6.9162e-04, 5.3112e-02, 1.7504e-02, 5.0199e-02, 4.4688e-02,
5.5349e-02, 4.1430e-02, 3.0638e-01, 1.0052e-02, 3.4906e-03,
1.0946e-02, 1.5992e-02, 1.1413e-01, 8.4667e-02, 9.4568e-02,
6.1189e-02, 6.7859e-08, 1.2937e-02, 3.1919e-02, 1.5799e-02,
3.9990e-02, 1.0280e-01, 3.3596e-02, 5.3143e-02, 8.9045e-02,
2.5200e-02], device='cuda:0')), ('layer1.1.conv1.weight', tensor([[[[ 2.5038e-02]],
[[ 2.5615e-03]],
[[ 1.0989e-02]],
...,
[[ 3.5722e-04]],
[[ 2.0726e-02]],
[[-3.4464e-02]]],
[[[ 2.2667e-03]],
[[-1.1549e-03]],
[[-1.1125e-02]],
...,
[[-1.6633e-02]],
[[ 2.2977e-02]],
[[-7.3551e-02]]],
[[[-6.0413e-03]],
[[-1.8692e-03]],
[[ 8.4820e-03]],
...,
[[-6.0168e-03]],
[[-3.8597e-02]],
[[ 2.8177e-02]]],
...,
[[[ 5.9032e-03]],
[[ 1.8610e-02]],
[[ 1.3279e-03]],
...,
[[ 1.0600e-02]],
[[-1.5613e-02]],
[[-4.9641e-02]]],
[[[ 4.5348e-02]],
[[-5.2813e-03]],
[[-1.3080e-02]],
...,
[[ 2.1859e-02]],
[[ 3.6713e-02]],
[[-2.1189e-02]]],
[[[ 6.5799e-03]],
[[ 1.9079e-02]],
[[ 1.4258e-02]],
...,
[[-3.7024e-03]],
[[-2.5108e-02]],
[[ 7.2490e-02]]]], device='cuda:0')), ('layer1.1.bn1.weight', tensor([ 1.5418e-01, 3.2359e-01, 2.4792e-01, 2.1051e-01, 1.7820e-01,
1.8067e-01, 1.7301e-01, 2.4763e-01, 1.7234e-01, 3.1702e-01,
1.4758e-01, 1.9952e-01, 1.2419e-01, 1.2985e-09, 1.6273e-01,
1.9145e-01, 1.6043e-01, 1.9408e-01, 2.5509e-01, 1.7340e-01,
1.4120e-01, 1.7380e-01, 1.7927e-01, 1.2663e-01, 1.6290e-01,
1.0250e-01, 1.7276e-01, 1.5056e-01, 1.7809e-01, 2.2617e-01,
2.8132e-01, 2.1156e-01, 2.3651e-01, 1.8941e-01, 1.6280e-01,
1.5587e-01, 1.6888e-01, 2.4972e-01, 3.4105e-01, 1.3929e-01,
3.0862e-01, 3.0111e-01, 1.7055e-01, 2.9651e-01, 1.1900e-01,
4.6577e-07, 1.7862e-01, 1.4380e-08, 1.4639e-01, 1.9021e-01,
1.9271e-01, 2.1601e-01, 1.5010e-01, 2.4548e-01, 1.3231e-01,
2.9485e-01, 8.5831e-09, 1.7127e-01, 1.3440e-01, 1.4374e-01,
2.1540e-01, 1.3796e-01, 1.9064e-01, 1.8207e-01], device='cuda:0')), ('layer1.1.bn1.bias', tensor([ 5.7919e-02, -2.0846e-01, -1.0967e-01, -1.0636e-02, -5.2057e-02,
-1.7326e-02, -5.5401e-03, -2.0876e-01, 8.0133e-03, -4.0513e-01,
2.5865e-02, -9.4683e-02, 9.5730e-02, -9.1670e-08, 9.9355e-03,
-4.4990e-02, 5.5640e-02, 2.2149e-02, -1.4557e-01, -2.4153e-01,
-6.7984e-03, 8.1301e-02, 9.7942e-03, 2.3542e-02, 1.0494e-01,
7.5696e-02, -1.4552e-03, 2.3584e-01, -4.5816e-02, -4.4147e-02,
-1.7976e-01, 1.5807e-03, -1.0492e-01, -9.8615e-02, 1.9448e-02,
-1.0210e-01, 2.7732e-01, -1.4269e-01, -2.2826e-01, 1.4129e-02,
-1.9629e-01, -2.3039e-01, -1.0216e-01, -1.9040e-01, 2.9462e-01,
-2.2354e-06, -1.1618e-01, -1.0213e-07, 1.0679e-01, -2.5207e-01,
-6.5190e-02, -7.5232e-03, 1.4403e-01, -7.3785e-02, 2.8200e-01,
-1.9581e-01, -6.0297e-08, 3.9227e-01, 9.5571e-02, 1.1742e-01,
-1.6532e-02, 2.5024e-01, -1.6168e-01, 9.6112e-03], device='cuda:0')), ('layer1.1.bn1.running_mean', tensor([-3.2618e-02, 1.2344e-01, 2.9107e-02, 1.2438e-01, 1.8299e-01,
2.0425e-02, 7.9855e-02, 1.4353e-02, -1.6327e-01, 1.6754e-01,
-2.5995e-01, 9.0388e-02, 2.6071e-01, 5.4474e-09, -1.1275e-01,
4.4073e-02, 9.6783e-02, -9.8441e-05, 4.5181e-02, -5.1553e-03,
-7.1977e-02, -7.7443e-02, -1.4635e-01, 9.8842e-02, 7.9017e-02,
9.5060e-02, -4.8912e-02, -5.1104e-02, 1.6378e-02, 8.5298e-02,
1.0308e-01, 2.6660e-02, 2.2627e-01, 8.3080e-02, 4.8937e-02,
1.2527e-02, -7.8198e-02, 4.4054e-02, 4.6868e-02, 1.2069e-02,
1.6559e-01, -1.1663e-03, 4.3683e-02, 7.3447e-02, 1.8800e-02,
-4.9475e-07, -2.2904e-01, 2.5030e-08, -1.0087e-01, -6.0277e-02,
9.5511e-02, 1.3999e-01, -1.4337e-01, 5.3128e-02, 3.8468e-02,
-1.0834e-01, 1.4099e-08, 4.6285e-01, -2.2469e-01, 3.8027e-02,
4.5822e-02, -3.4168e-02, -6.3448e-02, -6.1156e-02], device='cuda:0')), ('layer1.1.bn1.running_var', tensor([ 1.4654e-02, 6.8266e-02, 4.0650e-02, 2.5988e-02, 2.8965e-02,
2.8060e-02, 1.4982e-02, 2.7596e-02, 2.3759e-02, 3.1358e-02,
1.4901e-02, 2.4203e-02, 3.2630e-02, 1.1081e-15, 1.6930e-02,
2.5883e-02, 3.8990e-02, 4.1960e-02, 3.4098e-02, 9.5989e-03,
1.4908e-02, 2.4923e-02, 2.4058e-02, 2.0091e-02, 2.1373e-02,
1.7733e-02, 2.6151e-02, 3.0960e-02, 1.6736e-02, 2.5418e-02,
6.3766e-02, 2.8603e-02, 2.8821e-02, 1.8055e-02, 2.8665e-02,
1.7144e-02, 5.1241e-02, 3.3716e-02, 6.0787e-02, 1.4163e-02,
7.6656e-02, 2.9424e-02, 1.6245e-02, 6.0360e-02, 1.2882e-02,
9.0179e-14, 8.5261e-03, 5.6659e-16, 1.8502e-02, 8.3572e-03,
1.5219e-02, 3.6184e-02, 1.2469e-02, 3.3631e-02, 1.6409e-02,
5.0340e-02, 4.5506e-16, 2.6385e-02, 1.7645e-02, 1.0090e-02,
3.6297e-02, 1.7708e-02, 1.3911e-02, 2.4973e-02], device='cuda:0')), ('layer1.1.conv2.weight', tensor([[[[ 1.5131e-02, -2.0844e-03, 1.3178e-02],
[ 9.9589e-03, -1.3180e-02, -1.4355e-02],
[ 2.0671e-03, -1.2831e-02, -2.8570e-03]],
[[-1.0579e-02, -3.2069e-03, -9.4352e-03],
[ 1.3472e-02, 3.1053e-02, 3.7212e-02],
[-1.3569e-03, -7.9770e-03, -1.2346e-02]],
[[ 3.5541e-03, 1.6511e-02, 1.0947e-02],
[ 8.5486e-04, 6.9477e-03, 7.7736e-03],
[-1.6510e-03, -6.9018e-03, 5.7594e-03]],
...,
[[-1.8107e-02, -3.8545e-02, -2.2902e-02],
[ 5.7928e-02, 5.5715e-02, 3.9585e-02],
[-3.1099e-02, -5.4442e-02, -3.3741e-02]],
[[ 1.1299e-02, 3.0649e-02, 8.2794e-03],
[ 9.0970e-03, 1.4344e-02, 1.9952e-02],
[-6.4701e-03, 8.5228e-03, -2.3314e-03]],
[[ 1.0523e-03, -9.7366e-03, 4.7732e-03],
[ 1.3268e-02, -1.5277e-02, -1.5127e-02],
[-8.2769e-03, -2.3385e-02, -4.4172e-03]]],
[[[-1.7524e-02, -1.1281e-02, 3.3239e-03],
[ 8.7153e-03, -1.3304e-02, -1.8601e-03],
[ 2.2107e-03, -8.4356e-03, -6.6629e-03]],
[[-3.0955e-03, -3.2096e-03, 1.7255e-02],
[ 6.2367e-02, 4.6211e-02, 4.6354e-02],
[ 3.3049e-02, 1.2192e-02, 2.4536e-02]],
[[ 1.0525e-03, -1.8479e-02, -1.2107e-02],
[ 9.8989e-03, 2.1735e-03, -2.9522e-03],
[ 3.5191e-04, 5.2870e-03, 4.9564e-03]],
...,
[[ 1.6884e-03, -1.2523e-02, -1.8005e-02],
[ 1.4441e-03, 5.8991e-03, -2.4711e-02],
[ 1.3471e-02, 2.0758e-02, 9.6534e-03]],
[[-4.8604e-03, -2.5021e-02, -4.2191e-02],
[ 1.3725e-02, -2.0680e-02, -4.8109e-02],
[ 2.4640e-02, 1.4972e-02, -3.0200e-02]],
[[-3.2639e-03, 1.7118e-03, -1.0876e-03],
[-9.8774e-03, -2.8971e-03, -3.1781e-03],
[ 1.8190e-03, -1.2580e-02, -7.6514e-03]]],
[[[-3.5578e-02, 3.4780e-03, 5.8688e-03],
[-4.1381e-02, 2.2893e-02, -1.1461e-02],
[-2.9374e-02, 2.5839e-03, 4.8404e-03]],
[[-3.3849e-02, -3.9511e-02, -4.7046e-02],
[-5.1498e-02, -6.5836e-02, -4.1163e-02],
[-3.0716e-02, -4.0902e-02, -3.2305e-02]],
[[-3.6906e-03, 1.0416e-02, -4.5497e-03],
[-7.9338e-03, -1.7565e-02, 1.8837e-02],
[-8.6528e-03, -2.1467e-02, 4.6485e-03]],
...,
[[-4.4951e-03, 1.6729e-02, -2.3512e-02],
[-1.5071e-02, 3.1607e-02, -1.1856e-03],
[ 2.8345e-04, 1.2184e-02, -1.0772e-02]],
[[ 1.3048e-02, -3.1117e-03, -2.6407e-03],
[ 1.4620e-02, -1.1034e-02, -1.8682e-02],
[ 8.3067e-03, -6.9058e-03, -8.1581e-03]],
[[-3.1262e-03, 2.2658e-03, -3.5499e-02],
[ 2.9468e-02, 3.6942e-02, -1.6511e-02],
[-2.1330e-02, -2.0283e-03, -3.8132e-02]]],
...,
[[[ 1.9015e-02, -1.9787e-03, 1.1898e-02],
[ 3.3850e-02, 2.9796e-03, 1.5825e-02],
[ 2.4014e-02, 1.9589e-02, 7.7777e-03]],
[[ 1.2416e-02, 8.4996e-03, 2.0058e-02],
[ 2.1967e-02, -2.2207e-04, 1.3904e-02],
[ 1.7036e-02, 1.0798e-02, 3.2501e-02]],
[[ 6.9128e-03, 2.4620e-02, 2.0987e-02],
[ 1.3580e-02, 1.8883e-02, 8.9205e-03],
[ 3.2908e-02, 4.2235e-03, -7.6568e-03]],
...,
[[-1.1066e-02, -6.8930e-03, -2.3867e-02],
[ 2.1260e-02, 2.1823e-02, -2.9148e-02],
[-2.7998e-02, -2.3903e-02, -4.1546e-02]],
[[ 1.1273e-02, -4.2481e-03, -1.8486e-02],
[-2.9172e-02, -3.2462e-02, 3.1805e-03],
[-1.5779e-02, 1.5057e-03, -3.1037e-04]],
[[ 1.7304e-02, -1.3856e-02, 8.2595e-04],
[ 1.4277e-02, -3.0786e-02, -1.0011e-03],
[ 4.8532e-03, 1.4432e-02, 4.1100e-02]]],
[[[ 4.5908e-02, 3.7129e-02, 1.2975e-02],
[ 2.1326e-02, 6.1570e-03, -1.9008e-02],
[ 9.4289e-03, -2.4564e-02, -1.0380e-02]],
[[ 2.7198e-03, 1.4472e-02, 1.0209e-03],
[ 1.7572e-03, 7.6154e-04, 4.8318e-03],
[ 6.8438e-03, 1.4269e-02, 4.6343e-03]],
[[-1.6090e-03, 3.8158e-03, 3.3782e-02],
[-1.4349e-03, 1.5522e-02, 6.1718e-02],
[ 3.3080e-02, 4.4800e-02, 3.2992e-02]],
...,
[[-1.1974e-02, 1.0980e-02, -4.0344e-02],
[ 9.9280e-04, 1.7168e-02, -2.6205e-02],
[-1.5580e-02, 4.5384e-02, -2.6620e-02]],
[[-7.5865e-03, 3.9624e-03, 1.0910e-02],
[-1.0516e-02, -5.6382e-03, 2.5779e-02],
[-9.3009e-03, 1.6905e-02, 2.5413e-02]],
[[ 2.1498e-02, 3.2472e-03, -2.0411e-02],
[ 3.4126e-02, -1.7670e-02, -3.2316e-02],
[-2.3196e-02, -2.6713e-02, -1.7770e-02]]],
[[[-3.3570e-02, -5.8592e-03, -4.9914e-02],
[ 4.4431e-03, -8.1181e-02, -1.0034e-02],
[-2.1210e-02, 1.2360e-02, -7.4776e-03]],
[[-3.4235e-02, 1.6652e-02, -1.4267e-03],
[-8.8491e-03, 6.0021e-02, 5.9335e-03],
[-2.1950e-02, -5.6134e-03, -1.8367e-02]],
[[ 1.2906e-02, 1.5449e-02, 7.6274e-03],
[ 1.5200e-02, 8.9299e-03, 1.6331e-02],
[ 2.3584e-02, 1.4231e-02, 1.4983e-02]],
...,
[[ 2.4148e-02, 8.2161e-03, -2.7140e-03],
[ 2.0715e-02, -8.5727e-02, 2.1260e-02],
[ 1.4628e-03, 2.1745e-02, 1.2310e-02]],
[[ 3.3777e-02, 4.1994e-03, 1.6696e-02],
[ 3.3634e-03, -8.5029e-02, 2.6096e-02],
[ 7.1902e-03, -1.0627e-02, 1.0069e-02]],
[[ 2.1614e-02, 1.3685e-02, -1.9401e-03],
[ 2.8485e-02, -2.2742e-02, -9.2074e-03],
[ 4.3265e-02, 1.0341e-02, 1.4337e-02]]]], device='cuda:0')), ('layer1.1.bn2.weight', tensor([ 0.3002, 0.1879, 0.1683, 0.1700, 0.2510, 0.1264, 0.1596,
0.1925, 0.1379, 0.1619, 0.1925, 0.2059, 0.2133, 0.1886,
0.1552, 0.1556, 0.2230, 0.2143, 0.1503, 0.2025, 0.1546,
0.1728, 0.1657, 0.1653, 0.1759, 0.2397, 0.1492, 0.1925,
0.1350, 0.1303, 0.2352, 0.2116, 0.2162, 0.1721, 0.1860,
0.2056, 0.1355, 0.2451, 0.1831, 0.2053, 0.1650, 0.1824,
0.2217, 0.1820, 0.2021, 0.1699, 0.2040, 0.1753, 0.1816,
0.1904, 0.1531, 0.2186, 0.1661, 0.1781, 0.1438, 0.2048,
0.1353, 0.1997, 0.2582, 0.1683, 0.1852, 0.1189, 0.1655,
0.1964], device='cuda:0')), ('layer1.1.bn2.bias', tensor([-0.3349, -0.1008, -0.0095, 0.0479, -0.0964, 0.3131, 0.0501,
-0.0227, 0.2577, 0.1367, -0.1260, -0.0017, -0.1620, -0.0494,
0.1925, 0.0733, -0.0646, -0.1330, 0.0037, -0.1173, 0.0661,
-0.1220, 0.0396, 0.0579, 0.0875, -0.2095, 0.2687, -0.1505,
0.2573, 0.2496, -0.0825, 0.0500, -0.0324, -0.0489, 0.0205,
0.0002, 0.1885, -0.0236, -0.0888, -0.0540, -0.2141, -0.0731,
-0.0511, -0.0304, -0.0897, -0.0896, -0.1392, 0.0069, -0.1228,
-0.1339, 0.0683, -0.0021, -0.0092, -0.0254, 0.2965, 0.0168,
0.1691, -0.0496, -0.1790, 0.1099, 0.0447, 0.1081, -0.0762,
-0.0343], device='cuda:0')), ('layer1.1.bn2.running_mean', tensor([-0.1651, -0.0358, -0.0611, -0.1108, 0.0933, 0.0135, 0.0522,
-0.1282, -0.1551, -0.0586, -0.1046, 0.0876, 0.0530, 0.0465,
0.0862, 0.0631, 0.0286, 0.0615, -0.3105, -0.0605, 0.0084,
-0.0231, -0.0687, 0.0231, -0.0741, 0.0552, -0.1116, 0.0057,
-0.1035, 0.0832, 0.0336, -0.0038, -0.0631, -0.1846, -0.0248,
0.0074, -0.1125, 0.1376, -0.0949, 0.0031, -0.0579, -0.0792,
0.0801, -0.1601, -0.0412, 0.0193, -0.0757, -0.0211, -0.0768,
-0.0239, -0.1087, -0.0930, -0.2758, 0.0511, -0.1115, 0.0184,
-0.1497, -0.0629, 0.1135, -0.0199, -0.0436, -0.0387, -0.1013,
-0.0535], device='cuda:0')), ('layer1.1.bn2.running_var', tensor(1.00000e-02 *
[ 0.4983, 1.3560, 1.3164, 1.7145, 1.7537, 1.2157, 1.1327,
1.6260, 1.2436, 2.0826, 1.1711, 1.9701, 1.4177, 1.1659,
1.8687, 1.2662, 1.8831, 1.3339, 0.9580, 1.4539, 1.3261,
1.0986, 1.1786, 2.0392, 1.5480, 1.1363, 2.4024, 0.6805,
0.9819, 1.6173, 1.1828, 1.7875, 1.4325, 0.8746, 2.4996,
1.4306, 2.1290, 1.9596, 0.9103, 0.9728, 0.7097, 1.3366,
1.4355, 0.8063, 1.0785, 1.5200, 1.1027, 1.9080, 0.7931,
1.3609, 1.0939, 2.1999, 1.1185, 1.1921, 1.2707, 1.8775,
1.6848, 1.4180, 1.7031, 1.8952, 2.1918, 1.0665, 0.6919,
0.9480], device='cuda:0')), ('layer1.1.conv3.weight', tensor([[[[ 6.2414e-02]],
[[-3.7548e-03]],
[[-9.5037e-04]],
...,
[[-5.4637e-03]],
[[ 2.2708e-03]],
[[-7.4026e-03]]],
[[[ 9.6117e-02]],
[[ 7.1861e-03]],
[[-3.0478e-03]],
...,
[[-6.0525e-04]],
[[-1.9534e-03]],
[[-3.2167e-03]]],
[[[ 5.0736e-02]],
[[-2.1512e-02]],
[[-1.5421e-02]],
...,
[[ 9.1875e-03]],
[[ 8.9473e-03]],
[[ 2.6045e-02]]],
...,
[[[ 6.7911e-02]],
[[-1.1126e-03]],
[[-5.5242e-04]],
...,
[[-4.6133e-04]],
[[ 5.2510e-03]],
[[-3.8238e-03]]],
[[[-1.9320e-02]],
[[ 6.4255e-03]],
[[ 2.0596e-02]],
...,
[[ 1.0755e-03]],
[[-5.6299e-03]],
[[-1.3803e-02]]],
[[[ 4.5834e-02]],
[[-5.5471e-02]],
[[ 1.1134e-03]],
...,
[[-4.7031e-03]],
[[-1.9826e-02]],
[[-3.8611e-02]]]], device='cuda:0')), ('layer1.1.bn3.weight', tensor([ 2.1259e-03, -6.9715e-03, 9.4608e-02, -1.1439e-02, 1.4931e-01,
1.6228e-03, 1.9745e-01, 1.8000e-02, -1.7183e-04, 1.1793e-01,
1.0572e-02, 1.5288e-01, 6.2277e-03, -9.6050e-04, 1.0127e-01,
1.4762e-01, 1.1484e-01, 2.5359e-02, 9.5329e-02, 2.8729e-03,
7.1651e-02, 2.7586e-03, 1.5039e-01, 2.7510e-01, 9.0146e-02,
1.2768e-01, 5.5667e-03, 1.3082e-01, -1.9346e-03, 5.0423e-02,
2.1226e-03, -7.5048e-03, 1.8880e-02, 1.1878e-01, 9.8673e-02,
7.3502e-02, 1.5206e-01, 8.6612e-02, 5.0623e-03, -1.2326e-03,
1.1257e-01, 5.9433e-02, 1.4774e-01, 8.9319e-02, 9.7549e-02,
6.0202e-02, 8.6732e-02, 1.5079e-01, 8.0304e-02, 2.7696e-02,
-1.7456e-02, 1.7293e-01, 6.8518e-02, 1.4533e-01, 4.0339e-03,
2.5851e-02, 1.8742e-01, 6.8096e-02, 1.0810e-01, 1.1762e-01,
5.3549e-03, 7.9646e-02, -2.2026e-03, 1.5807e-01, 7.9670e-08,
3.1757e-05, 1.1837e-01, 3.6327e-02, 1.5932e-01, 3.8934e-02,
1.2485e-01, 6.9955e-06, 4.1722e-02, -1.8706e-03, 2.6218e-02,
1.5290e-01, 2.1404e-01, 1.1172e-01, -4.6150e-03, 1.1622e-01,
8.1356e-02, 4.0980e-02, 4.7201e-02, 4.9864e-02, 9.7825e-02,
7.0251e-02, 1.0111e-01, 7.0747e-02, -7.6863e-05, 3.2789e-04,
2.4123e-03, 1.1639e-01, 2.5553e-03, 1.1059e-01, 2.2347e-01,
1.6373e-01, 5.5450e-03, 1.1086e-01, 1.6115e-01, 1.3622e-01,
1.0025e-01, 1.9535e-06, 7.9823e-02, 4.2133e-03, 2.5824e-02,
1.2685e-01, 1.5052e-01, 9.5539e-02, 7.1596e-04, 6.7004e-02,
5.2219e-03, 5.9135e-02, 1.8622e-03, 5.1936e-02, 1.3801e-01,
-1.0016e-02, 2.3089e-03, 1.5128e-01, 5.8505e-02, 1.8369e-03,
1.3944e-01, 1.8065e-01, 1.2049e-01, -2.3750e-05, 1.6272e-01,
5.0985e-02, 1.4397e-01, -4.7397e-05, 7.9884e-02, 5.2822e-02,
1.1337e-01, 1.5378e-01, 1.0323e-01, 1.5092e-01, 8.5652e-02,
1.3144e-01, 8.7746e-02, 1.4908e-01, 1.1047e-01, 1.5284e-01,
1.2085e-01, 8.5849e-03, 5.9879e-02, 1.0519e-01, 5.8930e-02,
7.4151e-02, -4.9627e-03, 1.2280e-01, -1.1367e-01, 7.9317e-02,
1.2471e-02, 7.1211e-02, 2.1419e-01, 5.7342e-02, 7.2397e-02,
1.8971e-01, 1.3408e-01, 2.0415e-01, 2.4954e-03, 1.2848e-03,
8.5444e-04, 6.0994e-02, 1.9005e-03, 3.4890e-03, 1.7403e-01,
1.9638e-01, 1.8264e-01, 5.0056e-03, 1.0441e-01, 1.1164e-01,
2.0009e-01, 1.5155e-02, 1.8996e-01, 1.4341e-01, 1.5146e-01,
8.5111e-02, 8.0783e-02, 1.6485e-03, 1.0222e-01, 1.4608e-01,
-5.9042e-02, -1.8772e-03, 1.3604e-01, 1.9610e-01, 2.5466e-03,
2.1734e-06, -6.8739e-03, 4.4865e-03, 5.7102e-02, -5.8736e-03,
1.7767e-01, 9.2354e-04, -3.2072e-03, 2.2485e-02, 2.0056e-01,
4.5350e-03, 9.2664e-02, 8.6701e-02, 6.1892e-02, 1.6515e-01,
1.3486e-02, 8.1960e-02, 1.8972e-06, 2.1170e-02, 1.7591e-01,
-2.7518e-04, 1.5886e-01, 5.5978e-02, 7.8433e-02, 9.7138e-02,
8.8362e-02, -5.5156e-02, 1.8277e-01, 1.0541e-01, 1.7125e-01,
2.3317e-02, 1.0716e-01, 1.0549e-01, 7.3666e-02, 7.0549e-02,
3.6463e-03, 8.2517e-02, 1.3419e-01, -2.9456e-03, 1.0492e-01,
1.1982e-01, 1.6136e-04, 5.0766e-02, 1.9301e-03, 1.5054e-01,
1.1902e-01, -2.0978e-03, 1.9658e-01, 4.5569e-03, 6.8229e-02,
1.3388e-02, 1.8132e-01, -5.9605e-04, 1.6247e-01, 3.2930e-02,
1.1200e-01, 1.9393e-01, 2.5808e-03, 8.3123e-02, 1.1253e-01,
1.7063e-01, 1.0440e-01, 1.7874e-01, 7.7854e-02, 1.5666e-01,
5.0905e-02, -2.7906e-03, 6.0902e-03, 8.2056e-04, 2.6993e-02,
2.0848e-01], device='cuda:0')), ('layer1.1.bn3.bias', tensor([ 3.9980e-03, 5.0117e-03, -6.5528e-02, 1.8528e-03, -4.9248e-02,
2.5485e-03, 6.3341e-02, -6.2315e-03, 2.3204e-03, 4.1033e-02,
-1.3602e-02, 9.2113e-02, 5.1501e-02, 8.3470e-03, -3.8014e-02,
2.8642e-02, -3.2982e-02, -3.1293e-03, -9.6352e-02, 2.4645e-03,
1.8199e-02, 1.1311e-04, 6.3481e-02, 1.9443e-02, -3.1420e-02,
-5.5623e-02, 5.3881e-03, 1.0326e-02, 3.8256e-03, -3.4432e-02,
-1.9697e-03, 7.8351e-03, 2.7125e-02, 1.1254e-01, 9.4352e-02,
1.0976e-01, -1.0558e-01, -2.2601e-02, 3.3063e-03, 3.2106e-03,
-2.4838e-02, -2.3685e-02, -3.2189e-02, -7.1171e-02, 3.1299e-02,
4.1259e-02, 2.1263e-02, -6.1532e-02, -1.1498e-01, 5.9384e-02,
1.2950e-02, 9.2789e-02, -4.9363e-02, 7.3796e-03, 1.7671e-03,
-4.1717e-03, 1.4550e-01, -4.8883e-02, -7.7205e-02, -7.8253e-04,
9.3112e-03, -4.0488e-02, 3.3157e-03, 6.0696e-02, -2.8872e-07,
5.0370e-03, -6.1416e-02, 3.6578e-02, -6.5839e-02, 5.3615e-03,
-4.3997e-02, -5.9013e-05, 4.0512e-02, 4.0043e-03, -8.4623e-03,
7.2705e-02, 4.0860e-02, -5.0787e-02, 6.0637e-03, -1.8153e-02,
2.0342e-03, 5.1603e-02, 8.6173e-03, 9.7429e-02, 1.2427e-01,
9.1219e-02, 5.4813e-02, -2.4722e-02, 8.3121e-03, -2.7292e-03,
1.1337e-03, 9.3664e-03, 1.9196e-03, 6.6088e-03, 9.5043e-02,
5.2576e-02, 5.9158e-03, -5.3450e-02, 1.0382e-01, 1.3383e-01,
7.6762e-03, -3.7883e-05, -8.0345e-04, 2.2704e-03, 6.0525e-03,
-3.0990e-02, -1.4757e-01, -3.1099e-02, 1.7915e-03, 1.7390e-02,
2.5397e-03, 1.1629e-01, 2.4081e-03, 4.4117e-03, 5.3332e-02,
-6.1501e-03, 3.1353e-03, -3.3524e-03, 5.4908e-03, 3.5045e-03,
-7.7928e-02, -1.3230e-01, -6.2576e-02, -5.2930e-04, -1.7536e-02,
3.4290e-02, 2.3256e-02, -7.7662e-04, -7.5449e-02, 5.0531e-03,
-4.0019e-02, -1.0415e-02, 3.1160e-02, -1.5949e-02, -7.4926e-02,
-1.8892e-02, -1.6239e-02, -5.3514e-02, 5.5231e-02, -1.1096e-01,
-4.4315e-02, 2.6778e-03, -1.0696e-02, 1.6248e-02, -1.0240e-01,
-5.0648e-02, 1.3774e-02, 1.3104e-02, -3.1626e-02, 1.5090e-02,
-8.3022e-02, -4.2749e-02, -5.2607e-02, 3.0274e-03, 5.8953e-02,
5.3911e-02, -1.6867e-02, 9.5402e-03, 2.3234e-03, 3.7943e-03,
7.4951e-03, 2.6353e-04, 8.1738e-03, 2.3198e-03, 2.3122e-02,
3.4564e-02, -8.3066e-02, 9.4532e-03, -1.8433e-02, 2.1947e-02,
1.5602e-01, 6.9935e-03, -1.3894e-02, -2.8587e-02, -2.7557e-02,
-1.1292e-01, -2.8270e-02, 4.6501e-03, 1.9597e-02, 9.3797e-02,
4.3781e-03, -9.2987e-03, 3.9714e-02, 4.4164e-02, 3.5342e-03,
-3.3409e-06, -2.8622e-03, 6.1362e-02, -4.5803e-03, 4.8157e-03,
-5.8102e-02, 4.9529e-03, 3.6021e-03, -4.0624e-05, 1.9674e-02,
5.7639e-04, -4.6084e-03, 5.8968e-03, -3.6626e-02, -3.5292e-02,
-3.6024e-03, 3.8701e-03, -2.2053e-05, 1.0455e-02, -3.3503e-02,
3.2163e-03, -2.1732e-02, 6.6086e-02, 2.0005e-02, 1.3922e-01,
1.1378e-02, -6.7080e-02, 3.6496e-02, -2.4453e-02, -1.4834e-01,
7.1466e-03, -2.3019e-02, 1.9417e-02, 2.1519e-02, -2.5884e-02,
6.2051e-03, -3.8573e-02, 3.2588e-02, 6.4501e-03, 1.1272e-01,
-3.1017e-02, 4.9000e-03, -3.6400e-03, 7.3899e-03, 5.2943e-02,
-4.5050e-02, 6.7647e-03, 9.6522e-02, 3.5004e-03, 4.6488e-02,
-4.7906e-04, -1.1251e-01, 9.4231e-03, 3.7263e-02, -1.6908e-02,
-7.2099e-03, 1.2631e-01, 5.0302e-03, -4.8352e-02, 1.7477e-03,
-2.7385e-02, -7.2289e-02, -8.5673e-02, -5.9411e-02, 1.8720e-03,
1.2158e-02, 3.6557e-03, 6.4777e-03, 6.6302e-03, 2.9329e-02,
-7.0381e-02], device='cuda:0')), ('layer1.1.bn3.running_mean', tensor([-1.5714e-03, 3.4312e-03, 9.4634e-03, 1.0951e-03, -5.0856e-03,
-1.2469e-03, -3.9220e-02, 1.8221e-02, -3.8141e-03, 1.2045e-02,
-9.4127e-03, 8.4536e-02, -3.5008e-04, -1.2071e-03, -4.1583e-03,
9.1697e-02, -5.8518e-02, 5.2168e-03, 1.8181e-02, -1.9620e-03,
1.2106e-02, 1.1773e-03, -1.6550e-02, 1.4402e-02, -1.9154e-02,
3.6395e-02, -5.5203e-03, 1.0951e-02, 5.1018e-04, 5.2658e-03,
2.7055e-03, 3.6301e-03, 5.8333e-03, -1.3674e-02, 1.1719e-01,
-2.5244e-02, 4.0830e-02, -2.3513e-02, 6.4659e-03, 3.9299e-03,
-2.4468e-02, 2.8404e-02, -3.4214e-02, -2.6212e-02, 1.3743e-02,
-1.8971e-03, -5.8402e-03, -3.6669e-03, -5.2580e-03, -5.1293e-03,
-1.2480e-02, 8.3010e-02, 1.2319e-02, -1.5695e-02, -4.0148e-03,
-7.0125e-03, -5.5769e-02, -2.7152e-03, -5.6595e-04, -9.0007e-02,
-2.8079e-03, 2.4482e-02, 1.5382e-03, 4.7995e-02, 7.2283e-08,
1.1052e-03, -2.5143e-03, 7.3729e-03, -1.8615e-02, 7.1877e-03,
-1.9695e-02, -2.6035e-05, -2.2075e-02, -2.0722e-03, -2.0409e-02,
-6.1531e-02, 4.7805e-03, -1.3983e-02, 9.3593e-04, -4.0818e-03,
-1.7837e-02, -7.7156e-03, -6.3629e-03, 1.8930e-02, 7.3220e-03,
-4.6716e-02, -4.4801e-02, -4.0319e-02, -1.6370e-03, 1.6126e-03,
2.2437e-03, 1.2712e-05, -2.9303e-03, -7.5333e-02, 7.0419e-02,
8.3830e-02, -3.2950e-03, 4.0224e-02, 5.0944e-02, 3.4832e-02,
8.2021e-03, 6.6448e-06, -6.8727e-02, -2.2959e-04, -5.0971e-04,
2.7519e-02, 8.8991e-02, 1.1764e-02, -7.0619e-04, -2.0248e-02,
-1.1334e-03, -2.5409e-03, -1.8755e-04, -1.7704e-03, 8.1338e-02,
-1.1255e-04, 3.8242e-03, 1.9954e-02, -2.1407e-02, -9.7808e-03,
1.2224e-02, 9.6831e-02, 5.4322e-02, 2.2621e-05, 1.7543e-02,
2.5366e-02, 1.8658e-02, -5.5069e-05, 4.2827e-02, -7.6243e-02,
1.8431e-02, -3.5651e-02, -3.4577e-02, -7.6114e-02, -5.5688e-03,
1.9526e-02, -1.5677e-02, 3.9340e-02, 3.1463e-02, -4.5586e-02,
-2.8168e-02, -8.6423e-03, -2.8152e-02, 4.3470e-02, -5.1769e-03,
4.3537e-03, -2.5024e-02, -8.4987e-03, -7.9445e-02, -4.2511e-02,
5.2546e-03, -9.3598e-03, -3.0126e-02, -1.7505e-02, 2.6569e-02,
1.2683e-03, 4.3523e-02, -3.5517e-02, 1.4636e-03, -9.0041e-04,
-3.9615e-03, -3.3801e-02, -2.5182e-03, 9.0325e-03, -6.0671e-02,
-1.7460e-02, 3.6662e-02, -2.0421e-03, 3.2906e-02, 1.8158e-02,
4.6860e-03, -3.4271e-03, 1.0359e-01, 2.8883e-02, -4.1314e-02,
-1.2809e-02, -1.6082e-02, -1.5449e-03, -4.4366e-02, -2.4183e-02,
2.3147e-02, 2.9840e-04, 8.0035e-03, -5.5176e-02, 2.6187e-03,
-1.0193e-06, -2.0066e-03, -1.3424e-02, -2.3990e-02, 4.9734e-03,
5.8676e-02, 3.0396e-04, -2.2239e-03, 3.0321e-03, -1.0465e-01,
-2.3676e-03, 6.6736e-03, -6.0079e-04, 5.1213e-02, 4.3537e-02,
-4.8556e-03, -1.7918e-02, 4.6367e-06, 1.1941e-02, 4.8477e-02,
2.2312e-03, -1.9364e-02, -4.3366e-02, 1.0707e-02, -3.3358e-02,
3.9306e-02, 1.3428e-02, 1.0146e-02, -7.7767e-02, 1.0377e-02,
1.6835e-02, 6.0866e-02, -2.1115e-02, 1.4841e-02, -2.1803e-03,
4.0192e-03, -4.5023e-02, 1.0453e-02, -5.1113e-03, -2.1306e-02,
2.2438e-02, 6.7562e-03, 2.6618e-02, -5.1027e-03, 1.1819e-02,
1.0344e-02, 5.7469e-03, 1.2326e-02, -1.8895e-03, -6.9661e-02,
-7.7080e-03, -2.6162e-02, 6.1833e-04, 1.6822e-02, -4.4767e-03,
9.2391e-03, 1.1024e-01, -4.3210e-04, 1.9544e-02, -3.5386e-03,
3.6337e-02, 6.4805e-03, -1.4615e-02, -4.8474e-02, 3.4913e-02,
-5.7116e-03, 9.2647e-03, 5.4073e-04, 2.8234e-03, 2.3086e-02,
-2.7694e-03], device='cuda:0')), ('layer1.1.bn3.running_var', tensor([ 1.6497e-04, 3.8362e-04, 6.0497e-04, 3.1963e-04, 1.7644e-03,
1.4226e-04, 4.9757e-03, 6.2076e-04, 1.6890e-04, 2.6498e-03,
2.2167e-04, 3.0988e-03, 2.1370e-04, 2.7113e-04, 2.0343e-03,
2.7422e-03, 3.0600e-03, 4.0244e-04, 1.3090e-03, 2.0990e-04,
1.0315e-03, 1.2443e-04, 2.6555e-03, 3.3469e-03, 4.8374e-03,
1.2871e-03, 1.2247e-04, 2.3372e-03, 1.6322e-04, 1.2881e-03,
1.8008e-04, 1.8270e-04, 5.3567e-04, 1.4126e-03, 2.1412e-03,
3.0327e-03, 3.3160e-03, 1.4751e-03, 4.6071e-04, 1.4993e-04,
2.1025e-03, 1.6132e-03, 2.6779e-03, 1.7175e-03, 2.8532e-03,
5.9292e-04, 3.3715e-03, 1.6028e-03, 3.6984e-04, 2.3264e-04,
2.3049e-04, 4.2972e-03, 4.7850e-04, 2.6422e-03, 1.3561e-04,
3.5601e-04, 4.4400e-03, 1.3118e-03, 2.6764e-03, 1.7558e-03,
1.5818e-04, 2.5239e-03, 8.9878e-05, 1.8283e-03, 8.9794e-15,
3.1824e-04, 1.2830e-03, 1.3622e-03, 1.3732e-03, 7.1271e-04,
1.0576e-03, 7.0714e-10, 1.8657e-03, 3.5575e-04, 6.6909e-04,
1.8467e-03, 3.4014e-03, 8.0552e-04, 1.2543e-04, 1.7943e-03,
1.8615e-03, 1.1286e-03, 5.6055e-04, 1.0220e-03, 2.9768e-03,
4.0252e-03, 1.8731e-03, 6.9592e-04, 2.4924e-04, 1.8348e-06,
5.3325e-04, 1.1014e-03, 1.5386e-04, 1.3741e-03, 3.5518e-03,
2.8723e-03, 1.2696e-04, 1.6250e-03, 2.4078e-03, 2.6069e-03,
2.1019e-03, 1.6873e-10, 1.3461e-03, 1.0274e-04, 5.2834e-04,
4.2979e-03, 9.0586e-04, 1.2615e-03, 8.0362e-05, 1.3978e-03,
3.2732e-04, 1.6326e-03, 1.6817e-04, 1.3000e-03, 2.5985e-03,
1.9760e-04, 1.9282e-04, 1.7672e-03, 2.8012e-03, 2.1634e-04,
2.4621e-03, 1.2796e-03, 2.4360e-03, 2.1004e-08, 7.1070e-03,
1.0444e-03, 1.9220e-03, 8.6668e-09, 1.3229e-03, 1.1233e-03,
3.5856e-03, 2.2088e-03, 1.7464e-03, 2.2353e-03, 5.6107e-04,
3.5704e-03, 1.0684e-03, 1.7465e-03, 4.1756e-03, 1.1674e-03,
1.5816e-03, 2.2824e-04, 6.6629e-04, 2.0570e-03, 5.7788e-04,
2.4727e-03, 3.7931e-04, 2.6859e-03, 1.6102e-03, 1.0392e-03,
2.0832e-04, 1.0335e-03, 2.9681e-03, 1.0659e-03, 9.4177e-04,
2.8718e-03, 4.1355e-03, 3.1028e-03, 1.3646e-04, 8.8089e-05,
2.0997e-04, 1.2515e-03, 1.2233e-04, 5.7967e-04, 1.8251e-03,
5.6033e-03, 2.8728e-03, 1.8991e-04, 2.0217e-03, 2.1076e-03,
4.5074e-03, 1.2566e-04, 2.1092e-03, 2.1070e-03, 1.4550e-03,
1.2944e-03, 1.4766e-03, 1.6744e-04, 1.7485e-03, 3.7121e-03,
9.2634e-04, 6.2235e-05, 2.3885e-03, 3.4888e-03, 1.8084e-04,
3.1471e-13, 1.8466e-04, 3.3865e-04, 1.2382e-03, 5.1805e-04,
5.9712e-03, 1.2543e-04, 3.8937e-04, 2.3870e-04, 2.8670e-03,
1.0501e-04, 3.4833e-03, 2.6276e-03, 1.0886e-03, 1.4486e-03,
5.2244e-04, 1.2915e-03, 6.2813e-11, 7.4178e-04, 2.2809e-03,
1.1491e-04, 1.7425e-03, 1.7044e-03, 1.4071e-03, 2.8954e-03,
1.2856e-03, 7.3337e-04, 1.8168e-03, 1.5946e-03, 7.9391e-04,
4.4118e-04, 1.4426e-03, 1.6475e-03, 1.6060e-03, 1.3126e-03,
9.9750e-04, 1.1744e-03, 1.9827e-03, 4.0575e-04, 2.0473e-03,
2.2809e-03, 3.4641e-04, 9.9494e-04, 3.2345e-04, 2.9388e-03,
1.0703e-03, 2.1647e-04, 6.5492e-03, 2.4570e-04, 1.3422e-03,
1.4444e-04, 2.9563e-03, 1.3247e-03, 2.0473e-03, 6.1947e-04,
1.4608e-03, 2.9458e-03, 4.3075e-04, 2.3218e-03, 3.0224e-03,
6.0871e-03, 1.0140e-03, 2.6065e-03, 9.1789e-04, 1.9706e-03,
9.9821e-04, 4.3559e-04, 1.7555e-04, 1.8852e-04, 9.4176e-04,
3.8851e-03], device='cuda:0')), ('layer1.2.conv1.weight', tensor([[[[ 5.7859e-03]],
[[ 1.3108e-03]],
[[-4.1067e-03]],
...,
[[-4.2579e-03]],
[[-4.1070e-02]],
[[-3.8527e-02]]],
[[[ 1.9597e-03]],
[[-1.0637e-02]],
[[ 1.8890e-02]],
...,
[[ 4.8657e-03]],
[[-1.1662e-01]],
[[ 7.1150e-02]]],
[[[-1.3912e-02]],
[[ 1.8223e-03]],
[[ 2.2519e-02]],
...,
[[-2.2903e-02]],
[[-1.1148e-02]],
[[ 7.2517e-03]]],
...,
[[[ 2.5342e-03]],
[[-3.2014e-03]],
[[-1.0234e-02]],
...,
[[ 1.2434e-02]],
[[-1.1588e-02]],
[[ 1.7599e-02]]],
[[[ 4.4654e-04]],
[[ 5.1545e-03]],
[[-2.1940e-02]],
...,
[[-1.4156e-02]],
[[ 6.5771e-02]],
[[-2.3287e-02]]],
[[[-2.4907e-03]],
[[-7.4796e-05]],
[[-3.3184e-02]],
...,
[[-9.5518e-03]],
[[ 2.8951e-02]],
[[-1.3137e-02]]]], device='cuda:0')), ('layer1.2.bn1.weight', tensor([ 0.1913, 0.1692, 0.0998, 0.1557, 0.2209, 0.2088, 0.1592,
0.1858, 0.1681, 0.1715, 0.2173, 0.1619, 0.1758, 0.2133,
0.2372, 0.1801, 0.2278, 0.1665, 0.1719, 0.1989, 0.1980,
0.1442, 0.1775, 0.2147, 0.1963, 0.1785, 0.1716, 0.1923,
0.1763, 0.2147, 0.1436, 0.1808, 0.1411, 0.2095, 0.1432,
0.2235, 0.1475, 0.2037, 0.1705, 0.1889, 0.1268, 0.2000,
0.2031, 0.1841, 0.1522, 0.1639, 0.2384, 0.1928, 0.1385,
0.1841, 0.1849, 0.1573, 0.2080, 0.1951, 0.2446, 0.1780,
0.1835, 0.1956, 0.1273, 0.2016, 0.1753, 0.1431, 0.2104,
0.1973], device='cuda:0')), ('layer1.2.bn1.bias', tensor([-0.0211, -0.0407, -0.0112, -0.1323, -0.1030, -0.0420, -0.0627,
0.0480, 0.0803, 0.1414, -0.1111, -0.1918, 0.0786, -0.0938,
-0.1518, 0.0853, -0.1570, 0.1497, 0.0868, -0.0677, -0.0525,
0.1388, 0.0777, -0.0986, 0.0222, 0.0800, 0.0634, 0.0124,
-0.0755, -0.1144, 0.0747, -0.0626, 0.1637, -0.0787, 0.1791,
-0.0868, 0.1491, -0.0757, 0.0105, -0.1133, -0.0680, 0.0583,
-0.0410, -0.0515, 0.0492, -0.0792, -0.1330, -0.0540, 0.0605,
-0.1405, -0.0211, 0.1586, -0.1310, -0.0853, -0.1872, 0.0526,
-0.0925, -0.0578, 0.0782, -0.1039, -0.0210, 0.0612, -0.1123,
-0.0781], device='cuda:0')), ('layer1.2.bn1.running_mean', tensor([-0.1200, -0.1022, 0.1423, -0.0015, 0.1505, 0.0301, 0.0381,
0.0028, -0.1898, 0.1534, 0.0209, 0.0663, -0.0679, 0.0070,
-0.0412, 0.0958, -0.5701, 0.0389, 0.0501, 0.1257, 0.1420,
0.0854, 0.0660, -0.0627, 0.0167, 0.2849, 0.0862, 0.0874,
-0.0499, -0.0769, 0.1412, 0.0828, 0.0710, -0.0781, 0.1688,
0.0492, -0.0871, 0.0224, 0.1005, 0.1423, 0.0059, 0.0551,
-0.1233, 0.0999, -0.0381, -0.0072, -0.1353, -0.0473, -0.2363,
-0.1054, -0.0694, 0.0177, -0.0446, -0.1395, -0.0545, 0.1206,
-0.0454, -0.0592, -0.2520, 0.0754, -0.0760, -0.0559, -0.0291,
-0.0205], device='cuda:0')), ('layer1.2.bn1.running_var', tensor(1.00000e-02 *
[ 1.9267, 1.2960, 2.8475, 0.5023, 2.0362, 3.5297, 0.7313,
2.4530, 2.5315, 2.3917, 1.9170, 0.4868, 3.1702, 3.1005,
2.0788, 2.8558, 0.8278, 4.0121, 2.1935, 2.3412, 2.1779,
1.3877, 2.2824, 1.4365, 2.9833, 3.1375, 2.5839, 2.8962,
2.1417, 1.2452, 1.4509, 1.9957, 1.9422, 2.0672, 1.8191,
3.3180, 1.7739, 1.5643, 1.6827, 2.2093, 0.5985, 3.3430,
2.3180, 1.7507, 2.0462, 0.4694, 2.1752, 1.8792, 1.8941,
0.8632, 1.8905, 3.6105, 0.9631, 1.9086, 1.8194, 3.4878,
1.0107, 1.8504, 1.6422, 1.7752, 1.7204, 2.0628, 1.4350,
1.9088], device='cuda:0')), ('layer1.2.conv2.weight', tensor([[[[ 3.3096e-02, -2.2837e-03, -2.6060e-02],
[-1.8888e-03, -1.4137e-02, -7.5973e-03],
[-4.9556e-02, 2.8547e-02, -1.2019e-02]],
[[ 2.0762e-02, -6.0752e-02, 1.7711e-02],
[ 2.2331e-02, -5.3168e-02, 2.7273e-02],
[-1.8588e-02, 7.2604e-03, -4.4545e-03]],
[[ 1.1293e-02, 4.2911e-03, 2.0791e-02],
[-2.3401e-03, -1.7693e-02, 1.8178e-02],
[-5.9822e-03, -9.6152e-03, 8.1386e-03]],
...,
[[ 1.3661e-02, -1.2209e-02, -4.6657e-03],
[-3.2961e-03, -3.7944e-03, -2.1736e-02],
[-2.4662e-02, 2.3986e-02, -1.5785e-02]],
[[-4.2326e-02, 5.5694e-03, 1.0661e-02],
[-4.7409e-02, 3.7776e-02, 2.1256e-02],
[-9.4837e-03, 1.1733e-02, -6.1717e-03]],
[[ 3.1075e-03, -7.9047e-03, -2.5325e-02],
[ 8.5359e-03, -6.7326e-03, -1.2789e-02],
[ 7.2580e-03, -1.6795e-02, -7.6375e-03]]],
[[[ 2.1006e-02, -1.9891e-02, -2.5320e-02],
[-3.3499e-04, 1.9618e-02, -2.0495e-02],
[ 4.6272e-02, 7.7210e-02, -5.5360e-02]],
[[-3.7223e-02, 6.7949e-02, -3.3836e-02],
[-5.4784e-02, 6.3341e-02, -3.2365e-02],
[-5.3404e-03, -1.1135e-02, -7.8994e-04]],
[[ 2.9878e-02, -5.3996e-03, 9.8225e-03],
[-3.4435e-03, 1.1870e-03, -4.5666e-03],
[-5.3928e-03, -9.0694e-03, 3.1196e-03]],
...,
[[-1.8036e-02, 2.4209e-02, 7.9403e-03],
[-8.6619e-03, 2.1570e-02, -6.4812e-03],
[-2.2654e-02, 8.9033e-03, 2.2195e-02]],
[[ 6.1954e-03, -1.9543e-02, -1.2755e-02],
[ 1.8484e-02, -1.3425e-02, 2.4133e-05],
[ 1.7827e-03, -5.4650e-03, 3.1103e-02]],
[[-1.4821e-02, 1.6666e-02, -3.1383e-02],
[-9.5354e-05, 2.7465e-02, 2.4956e-02],
[ 7.7412e-03, 2.5979e-03, -2.1601e-02]]],
[[[-1.4079e-02, 8.0381e-03, -1.7125e-02],
[ 3.8214e-03, -2.4952e-02, -2.6993e-02],
[ 2.8850e-02, 1.9032e-02, -8.6727e-03]],
[[ 8.9605e-03, 1.1186e-02, 2.7754e-02],
[ 3.4283e-03, -6.6164e-02, 2.5678e-02],
[ 3.7452e-02, -2.1241e-02, -2.8383e-02]],
[[ 1.2442e-02, 5.2476e-03, -5.9283e-03],
[ 1.4631e-02, -1.2374e-02, -2.8127e-03],
[-8.1462e-04, -2.5496e-03, -6.6706e-03]],
...,
[[-1.4014e-02, 2.1514e-02, 5.0287e-02],
[-6.9800e-02, -8.5586e-03, 2.6910e-02],
[-4.7790e-02, -6.6275e-02, 3.5433e-02]],
[[-1.2001e-03, -2.2680e-02, -7.8468e-03],
[-3.3417e-02, 2.9589e-02, -8.3864e-02],
[-1.2242e-02, 2.2470e-02, 8.2582e-03]],
[[-1.6228e-03, 1.9872e-02, -5.9169e-03],
[ 9.8616e-03, -4.3809e-03, 1.3927e-02],
[-5.0480e-03, 2.1110e-02, -2.3370e-03]]],
...,
[[[-4.9515e-02, -2.5277e-03, 2.3704e-03],
[-2.5449e-02, 2.5781e-03, 1.3212e-02],
[-2.8367e-02, 6.1904e-03, -7.1778e-04]],
[[-1.6230e-03, -1.3667e-02, 1.6481e-02],
[-7.6718e-03, -3.8440e-02, 4.1870e-02],
[-8.3919e-04, -2.0845e-02, 8.6331e-03]],
[[ 2.7622e-02, 9.3908e-03, 1.2196e-02],
[ 2.0812e-02, 2.6994e-03, -4.2113e-03],
[ 2.7357e-02, 7.6224e-04, -8.0598e-03]],
...,
[[-4.1468e-02, -2.5943e-02, 2.7411e-02],
[-6.6122e-02, -4.1655e-02, 5.9899e-02],
[-6.1420e-02, 2.5395e-02, 3.8950e-02]],
[[ 6.2085e-02, 5.7126e-03, 8.0286e-05],
[-1.9108e-02, -3.2557e-02, -3.4181e-02],
[-2.2809e-02, -9.4224e-03, -1.4776e-02]],
[[-2.6983e-02, -2.4507e-02, -1.3865e-02],
[-4.2552e-02, -3.6843e-02, 6.1647e-02],
[-3.8629e-02, -2.1726e-03, 4.2447e-02]]],
[[[ 7.4762e-03, -1.5865e-02, 2.5887e-02],
[ 2.6152e-02, -1.9164e-02, -6.5627e-03],
[-4.7026e-05, -1.0471e-02, -1.3037e-03]],
[[-4.9933e-03, 1.1487e-02, -3.9681e-02],
[-1.2797e-02, 1.7415e-02, -2.6448e-02],
[ 3.2297e-02, 1.2037e-02, -1.5810e-02]],
[[ 3.1153e-02, 4.1138e-03, 1.1098e-02],
[-2.8878e-03, -4.0853e-03, -4.8585e-03],
[ 4.2662e-03, -5.7329e-03, 7.8327e-03]],
...,
[[-2.5365e-02, 1.6538e-02, 5.1379e-02],
[-3.9035e-02, 3.4860e-02, 1.0149e-02],
[-4.4507e-02, 3.3185e-02, -2.4550e-02]],
[[-2.4078e-02, -1.9366e-02, -1.0191e-02],
[-1.6126e-02, 1.9683e-03, 5.2968e-02],
[-1.1998e-03, 2.4451e-02, 6.5782e-03]],
[[-5.2594e-03, -2.8607e-03, 2.9125e-02],
[ 1.5266e-02, 6.4225e-03, -9.4151e-03],
[ 3.6973e-02, -5.3924e-04, -2.3406e-02]]],
[[[ 9.4780e-03, 1.8678e-02, 4.8434e-03],
[ 2.7248e-02, -3.3991e-02, 1.4853e-02],
[-4.3406e-02, 1.9014e-02, -1.0946e-02]],
[[ 3.8813e-03, -1.0640e-02, -1.2859e-02],
[ 1.4223e-02, 1.7025e-02, 1.5361e-02],
[-2.6170e-02, -1.2333e-02, 3.2009e-02]],
[[ 1.1578e-02, 1.5274e-04, 3.5564e-03],
[ 7.9022e-03, 2.6997e-02, 9.6522e-03],
[-1.3545e-03, -2.9908e-04, 1.0236e-02]],
...,
[[-4.0883e-02, -1.2606e-02, 5.2883e-03],
[ 1.9850e-02, -3.1524e-02, -4.9819e-02],
[ 1.5679e-02, 4.7178e-02, -2.1810e-02]],
[[-1.1884e-02, -1.6430e-02, 4.3056e-03],
[ 2.1827e-02, -3.3460e-02, -1.5285e-02],
[-1.9582e-04, 3.6705e-02, -1.6792e-02]],
[[-1.6154e-02, 1.7052e-02, 1.9213e-03],
[-1.3728e-02, 2.2707e-02, 7.7133e-03],
[ 2.4167e-02, -9.8935e-03, -8.9031e-03]]]], device='cuda:0')), ('layer1.2.bn2.weight', tensor([ 0.2164, 0.2510, 0.2231, 0.2292, 0.2206, 0.2179, 0.2162,
0.1634, 0.2093, 0.2222, 0.2535, 0.3154, 0.2342, 0.2426,
0.2290, 0.2137, 0.2216, 0.1808, 0.2402, 0.2401, 0.2296,
0.2416, 0.2316, 0.2086, 0.2435, 0.2377, 0.2285, 0.2185,
0.2242, 0.2128, 0.2328, 0.2420, 0.1663, 0.2489, 0.1979,
0.2088, 0.2402, 0.1934, 0.1159, 0.2452, 0.2377, 0.1765,
0.2250, 0.1855, 0.2238, 0.2285, 0.2402, 0.2207, 0.2157,
0.2389, 0.2259, 0.2190, 0.2020, 0.2269, 0.2276, 0.2248,
0.2533, 0.1996, 0.2053, 0.1992, 0.2117, 0.2513, 0.2136,
0.2317], device='cuda:0')), ('layer1.2.bn2.bias', tensor([-0.0306, -0.0923, -0.0414, -0.0653, -0.1028, -0.0986, -0.0136,
-0.0025, -0.1152, -0.0291, -0.0657, -0.3410, -0.0687, -0.0501,
-0.0541, -0.0482, -0.0685, 0.0415, -0.0591, -0.0672, -0.0274,
-0.1126, -0.0099, -0.0306, -0.1327, -0.0547, -0.0480, -0.0265,
-0.0448, -0.0258, -0.0761, -0.1036, 0.0966, -0.0771, 0.0057,
-0.0256, -0.1545, 0.1036, 0.2601, -0.1233, -0.0618, 0.0686,
-0.0530, 0.0247, -0.0409, -0.0814, -0.0808, -0.0434, -0.0131,
-0.0899, -0.0204, -0.0730, -0.0149, -0.0785, -0.0759, -0.1011,
-0.0613, -0.1066, 0.0035, 0.0221, -0.0881, -0.0364, -0.0522,
-0.0728], device='cuda:0')), ('layer1.2.bn2.running_mean', tensor([-0.0098, -0.0400, -0.0904, -0.0330, -0.1189, 0.0177, -0.0773,
-0.1255, -0.0588, 0.0493, 0.0156, -0.1197, -0.0351, -0.0507,
-0.0374, -0.0654, -0.0070, -0.0527, -0.0139, -0.0015, -0.0679,
-0.1320, 0.0064, -0.0111, -0.0633, -0.0019, -0.0339, -0.0177,
-0.0152, -0.0266, -0.0192, -0.0388, -0.0239, 0.0171, -0.0317,
-0.0674, -0.1708, -0.0292, 0.0072, -0.0105, -0.0340, -0.1683,
0.0755, -0.0268, -0.0282, 0.0229, -0.0619, -0.0591, 0.0445,
-0.1075, -0.0686, -0.0541, -0.0706, -0.0176, -0.0124, -0.0563,
-0.0131, -0.0296, -0.0264, 0.0011, 0.2363, -0.0184, -0.0211,
-0.1095], device='cuda:0')), ('layer1.2.bn2.running_var', tensor(1.00000e-02 *
[ 2.0991, 3.2251, 1.7828, 1.7505, 1.1659, 1.4577, 1.6746,
0.7857, 1.5247, 2.0339, 2.8517, 0.4571, 2.4881, 2.4000,
2.4448, 2.2274, 1.8414, 1.7511, 2.0073, 2.4138, 2.1527,
1.1997, 2.8421, 2.6478, 1.3077, 2.6120, 2.5241, 2.2160,
2.2076, 1.9652, 2.2142, 1.9035, 1.1297, 2.9932, 1.6067,
1.5167, 1.4424, 1.8639, 1.1963, 1.2640, 2.9638, 1.3193,
1.7187, 1.5423, 2.6927, 2.3139, 2.0377, 1.8759, 2.2134,
1.6304, 2.2084, 1.6159, 2.1488, 1.9414, 1.9175, 1.7524,
2.4317, 1.1449, 1.8541, 1.6722, 1.1185, 2.5661, 2.0293,
1.3209], device='cuda:0')), ('layer1.2.conv3.weight', tensor([[[[ 4.2877e-03]],
[[-2.7595e-03]],
[[-1.9950e-03]],
...,
[[ 2.0631e-03]],
[[ 5.3560e-03]],
[[ 2.3445e-03]]],
[[[ 1.1497e-03]],
[[ 3.0026e-03]],
[[ 6.7865e-03]],
...,
[[-2.5102e-03]],
[[ 8.1134e-03]],
[[-6.5756e-03]]],
[[[-1.9855e-02]],
[[-1.0414e-02]],
[[-6.7570e-02]],
...,
[[-2.4544e-02]],
[[-5.9988e-03]],
[[-1.2558e-01]]],
...,
[[[-1.2564e-03]],
[[ 2.2247e-04]],
[[ 5.4108e-03]],
...,
[[ 6.6900e-04]],
[[ 8.4452e-03]],
[[ 3.9342e-03]]],
[[[ 2.4323e-04]],
[[ 3.5110e-03]],
[[-5.8671e-03]],
...,
[[-5.1452e-03]],
[[-4.9488e-03]],
[[-2.9513e-03]]],
[[[-4.7445e-02]],
[[-3.2359e-02]],
[[ 8.7406e-03]],
...,
[[ 9.7492e-03]],
[[-2.9358e-02]],
[[-3.6021e-02]]]], device='cuda:0')), ('layer1.2.bn3.weight', tensor([-0.0016, 0.0039, 0.2425, 0.0131, 0.2001, -0.0003, 0.1157,
0.0469, -0.0015, -0.0213, -0.0027, 0.1475, 0.2144, -0.0070,
0.1194, 0.1255, 0.0041, 0.0799, -0.0005, -0.0005, 0.1292,
0.0793, 0.1562, 0.1081, 0.0817, 0.2245, 0.0016, 0.0518,
-0.0021, -0.0041, 0.0056, 0.0016, -0.0075, 0.0553, 0.1322,
-0.0045, 0.0022, -0.0048, -0.0034, 0.0023, 0.3502, 0.0166,
0.1343, 0.0087, -0.0022, 0.2173, 0.0004, 0.1888, 0.2450,
0.1921, -0.0156, 0.0434, 0.2315, 0.1878, 0.0056, 0.0068,
0.0010, 0.1202, 0.0016, 0.0038, 0.0165, 0.0038, -0.0037,
0.0034, 0.3755, 0.0010, 0.2073, 0.0140, 0.0614, 0.0004,
0.1475, 0.2578, 0.0643, -0.0057, -0.0029, 0.0920, 0.1271,
0.0208, -0.0058, 0.0505, 0.0009, -0.0069, 0.0040, 0.2250,
0.1074, 0.1481, 0.1807, 0.2502, -0.0038, 0.2358, -0.0017,
0.0852, 0.0010, 0.2784, 0.1446, 0.0625, 0.0034, 0.1425,
0.0004, 0.0504, 0.0321, 0.2767, 0.1396, 0.0061, 0.0592,
0.0856, 0.1206, 0.2202, 0.0017, -0.0057, -0.0030, 0.0022,
0.0006, 0.0117, 0.0708, 0.0683, 0.0196, 0.1942, 0.1436,
-0.0469, 0.0319, -0.0979, 0.0079, 0.3327, -0.0042, 0.0008,
0.1136, 0.3380, 0.1424, 0.0981, 0.1345, 0.0581, 0.1317,
0.2326, 0.1753, -0.0130, 0.1274, 0.1846, 0.0102, 0.1972,
0.1870, -0.0045, 0.0482, 0.1322, -0.0003, 0.0024, 0.0184,
0.0478, 0.0400, 0.1090, 0.0005, 0.1481, 0.1863, 0.1207,
0.0002, 0.1088, 0.0022, 0.1066, 0.0067, -0.0003, 0.0016,
0.0525, 0.0004, 0.0073, 0.1850, 0.0008, 0.1266, 0.0059,
0.0487, 0.2174, 0.0690, 0.0029, 0.1675, 0.2123, 0.1973,
0.1488, 0.0617, 0.0012, 0.1072, 0.1278, 0.2105, 0.0081,
0.0629, 0.0711, -0.0027, 0.1726, 0.0000, 0.0759, 0.0108,
0.0019, 0.0082, -0.0008, 0.0039, -0.0305, 0.1661, -0.0016,
0.0711, 0.0164, -0.0029, 0.1796, 0.0242, 0.0327, 0.2187,
0.0048, 0.1950, 0.0014, 0.1541, 0.1010, 0.0469, 0.0230,
0.1238, 0.2172, 0.1265, 0.0617, 0.0626, 0.0042, 0.1738,
0.1467, 0.0994, 0.0942, -0.0024, 0.1659, 0.1885, 0.0002,
0.0056, 0.0617, -0.0044, 0.0083, 0.0039, 0.0799, 0.2456,
-0.0040, 0.0625, 0.0030, -0.0019, 0.0058, 0.1215, -0.0049,
0.0302, 0.0103, 0.0873, 0.0342, -0.0005, -0.0038, 0.0264,
-0.0051, 0.2376, 0.1021, 0.1778, 0.1523, -0.0035, 0.0004,
-0.0009, -0.0003, -0.0130, 0.1728], device='cuda:0')), ('layer1.2.bn3.bias', tensor([ 0.0017, 0.0050, -0.0054, 0.0124, -0.0881, 0.0005, -0.0247,
-0.0418, 0.0035, -0.0235, 0.0070, -0.1078, 0.0535, 0.0057,
-0.0897, 0.0597, 0.0052, -0.0405, -0.0603, 0.0030, -0.0492,
-0.0297, -0.0631, -0.0590, -0.0582, -0.0576, 0.0020, 0.0795,
0.0040, 0.0065, -0.0154, 0.0020, 0.0045, 0.0028, -0.0064,
0.0042, 0.0097, 0.0084, 0.0023, 0.0032, -0.0188, 0.0164,
-0.1280, 0.0127, 0.0017, 0.0174, 0.0038, -0.0713, 0.0007,
0.0770, 0.0008, 0.0517, 0.1328, -0.0958, 0.0005, 0.0035,
0.0027, -0.0058, 0.0045, 0.0072, 0.0020, 0.0050, -0.0005,
0.0033, -0.0338, 0.0036, 0.0159, 0.0020, -0.0928, 0.0044,
-0.0422, 0.0695, -0.0106, 0.0024, 0.0076, -0.0066, 0.1455,
-0.0577, 0.0016, -0.0517, 0.0034, 0.0016, 0.0090, 0.0579,
0.0588, 0.0985, -0.0513, 0.0631, 0.0069, 0.1254, 0.0027,
-0.0781, 0.0034, -0.0169, -0.1071, -0.0529, 0.0066, -0.1662,
-0.0032, 0.0389, -0.0090, 0.0724, -0.1217, 0.0030, -0.0496,
-0.0649, -0.0251, 0.1595, 0.0010, 0.0062, 0.0007, 0.0025,
0.0030, 0.0086, -0.1288, -0.0311, 0.0082, 0.0761, -0.0557,
0.0083, -0.0369, -0.1549, -0.0042, 0.0209, 0.0010, 0.0079,
-0.1325, -0.0068, 0.1754, -0.0212, -0.1287, -0.0491, -0.0283,
-0.0692, 0.1399, -0.0108, -0.0215, 0.0905, 0.0083, 0.1152,
0.0094, -0.0001, -0.0542, -0.0691, -0.0366, 0.0085, 0.0077,
-0.0154, -0.0710, 0.0421, 0.0052, -0.0696, -0.0764, 0.0922,
0.0041, -0.1396, 0.0024, -0.1090, 0.0033, 0.0044, 0.0022,
-0.0410, 0.0007, 0.0006, -0.0503, 0.0026, -0.1579, 0.0064,
-0.0523, -0.0934, 0.0422, 0.0069, -0.1208, -0.0240, -0.0178,
0.0259, -0.0662, 0.0054, -0.0769, -0.0720, -0.0847, -0.0050,
-0.0666, -0.1055, 0.0026, 0.1969, 0.0033, 0.0615, 0.0110,
0.0044, 0.0070, 0.0049, 0.0041, -0.0257, 0.0801, 0.0021,
-0.0666, 0.0033, 0.0020, 0.0068, -0.0105, -0.0078, 0.1224,
0.0045, -0.0717, 0.0086, -0.0488, 0.1549, -0.0310, 0.0254,
-0.0505, 0.0710, -0.0475, 0.0025, -0.0973, 0.0013, -0.0858,
-0.1063, -0.0699, -0.0341, 0.0053, -0.1138, 0.0241, 0.0034,
0.0031, -0.0646, 0.0038, 0.0027, 0.0046, 0.0062, -0.1204,
0.0027, -0.0078, -0.0002, 0.0036, 0.0062, -0.1684, 0.0071,
-0.0300, -0.0249, -0.0095, 0.0118, 0.0034, 0.0103, -0.0524,
0.0034, 0.0167, -0.1474, 0.0302, -0.0559, 0.0060, 0.0043,
0.0034, 0.0053, 0.0052, -0.1542], device='cuda:0')), ('layer1.2.bn3.running_mean', tensor([-1.0035e-03, -2.7422e-03, -9.5635e-02, -2.0726e-04, 4.8132e-02,
-1.0544e-03, -2.6496e-02, -2.4920e-02, -2.2136e-03, 2.7106e-03,
-2.4910e-03, -8.3024e-03, -6.3298e-02, 1.6686e-03, -5.8166e-02,
3.5711e-02, -2.1728e-03, -4.4013e-02, 1.1632e-03, -2.2347e-03,
-3.2484e-02, -1.3597e-02, 5.1380e-03, 1.2175e-02, -5.2279e-02,
-2.2949e-02, 9.0200e-04, -3.9777e-03, 6.0040e-04, -2.7460e-04,
-2.4295e-03, -4.9252e-03, 1.7820e-04, -4.3161e-03, 3.7351e-02,
-2.8888e-03, -2.5165e-03, -1.1536e-03, -1.9066e-03, -3.1034e-03,
-4.4320e-02, -9.6475e-05, -5.4383e-02, 2.1638e-03, -1.0909e-03,
-1.4962e-02, 1.3113e-03, -9.4112e-03, -1.0706e-02, 6.2918e-02,
9.0950e-04, -3.1809e-03, -5.8750e-02, -2.6148e-02, -9.4866e-04,
2.6346e-06, 5.4713e-03, -2.2663e-02, -1.2359e-03, -1.8834e-03,
-1.2598e-02, -3.8876e-03, 5.2850e-04, -2.8322e-03, -1.3142e-01,
3.0902e-03, 2.3478e-02, -8.5304e-04, 1.4828e-02, -5.1246e-04,
-5.1301e-02, -1.5087e-02, -6.4656e-02, -1.5704e-03, -2.2568e-03,
-1.2967e-02, -2.0454e-02, 5.1967e-03, 1.8417e-03, -1.6389e-04,
-1.9921e-03, 2.6515e-03, -4.8134e-04, -6.9821e-02, 2.8866e-02,
-2.1267e-02, -2.5920e-02, -6.5970e-02, 4.1644e-03, 2.0526e-02,
5.2993e-03, -2.2196e-02, 1.3863e-03, -4.5024e-02, -5.1349e-03,
2.1795e-02, -1.0778e-03, 3.1961e-02, -7.2894e-03, -2.6358e-02,
9.4724e-04, -5.5236e-03, -4.9898e-02, 2.7488e-03, -1.7153e-02,
-2.7667e-02, 2.5742e-02, -9.0447e-02, -1.5212e-03, 3.4386e-03,
1.2187e-03, -1.9168e-03, -2.5168e-03, 3.7760e-03, -1.0156e-02,
-4.3533e-02, -1.6350e-03, -5.9779e-02, -5.2690e-02, 1.4681e-03,
-2.8697e-03, -9.6624e-03, 4.1621e-03, -8.0560e-02, 6.0275e-04,
5.4571e-03, 2.1320e-03, -7.7299e-02, -5.6486e-03, -6.9692e-02,
-7.3579e-02, -8.3873e-03, -4.8202e-02, -3.0844e-02, -4.3224e-02,
-1.1860e-03, -5.4718e-02, 6.5475e-02, -2.9044e-03, -6.0999e-02,
4.2386e-02, 1.4522e-03, -2.0386e-02, -1.8766e-02, 4.6532e-07,
3.1428e-03, -5.8148e-04, -2.5041e-02, 2.2790e-02, 3.3111e-02,
2.8610e-03, 4.8122e-02, 4.0667e-02, -8.4077e-03, -3.3271e-04,
-1.3843e-02, -3.8978e-03, 1.1267e-03, 6.6504e-04, -1.0866e-03,
2.1002e-03, -4.5715e-02, -2.2389e-03, -3.2652e-03, 1.6212e-02,
-2.1429e-03, 2.6450e-02, 2.2031e-04, -1.5721e-02, -4.6404e-02,
5.0086e-03, -4.4962e-04, 1.8260e-02, 1.1919e-02, 1.3430e-02,
-3.5134e-02, -3.8083e-02, 2.0333e-03, -1.5442e-02, -2.4658e-02,
-6.0285e-02, 6.1141e-04, -3.4456e-02, -3.4378e-04, -1.3072e-04,
-2.3707e-03, -3.1295e-03, 1.4769e-02, -4.4187e-03, -2.4274e-03,
-2.9567e-03, -2.4061e-03, 9.7445e-04, 1.4216e-02, 1.3157e-03,
-1.4089e-03, -2.5121e-02, -7.5043e-03, 1.6309e-03, 2.9274e-02,
-1.8100e-03, -9.0115e-03, -3.7842e-02, -1.2484e-03, 2.4038e-02,
1.2304e-03, -1.6038e-02, -8.6509e-03, -1.1457e-03, -1.2833e-02,
-1.9168e-02, -7.5671e-02, -7.1391e-03, -1.0614e-02, -1.2244e-02,
1.0107e-03, -1.2118e-02, -4.5587e-02, 2.3452e-03, -3.4006e-02,
1.9312e-04, -1.9821e-02, 2.5749e-02, 1.6633e-03, 5.2321e-04,
-8.6818e-03, -1.7339e-04, -1.4667e-03, -3.6860e-03, -2.8286e-02,
1.7243e-02, 1.9510e-03, -6.0811e-02, -9.8544e-04, -2.0042e-03,
2.3218e-03, -2.3745e-02, -8.6949e-03, -1.9071e-02, 1.6408e-03,
-2.3473e-03, 2.0621e-03, 1.8342e-03, -1.9647e-04, -1.2856e-02,
1.9656e-03, -5.6594e-03, -1.5089e-02, -9.9969e-04, 2.5561e-03,
-9.5519e-04, -9.0107e-05, 4.6836e-03, -3.2230e-03, 1.5311e-03,
-6.4266e-02], device='cuda:0')), ('layer1.2.bn3.running_var', tensor(1.00000e-03 *
[ 0.2359, 0.6420, 1.9628, 0.2069, 1.8125, 0.1652, 2.2126,
1.1254, 0.2539, 0.2379, 0.2557, 1.3024, 3.5319, 0.3369,
2.0310, 1.7374, 0.1388, 1.6883, 0.0643, 0.3258, 1.2682,
0.6881, 1.6780, 1.0023, 1.5958, 1.5880, 0.1561, 0.7169,
0.1908, 0.1281, 0.1419, 0.2139, 0.1901, 0.5104, 1.3917,
0.1193, 0.1703, 0.1453, 0.5643, 0.2949, 6.3701, 0.3764,
1.6297, 0.2071, 0.0691, 2.0057, 0.0968, 1.7348, 2.8310,
3.3054, 0.2294, 0.6591, 2.3799, 1.3003, 0.1272, 0.1640,
0.1734, 1.2311, 0.1380, 0.1126, 0.1695, 0.1565, 0.1158,
0.0678, 3.7054, 0.4702, 2.7974, 0.1893, 0.2936, 0.1430,
1.1352, 4.7637, 0.9724, 0.4972, 0.1902, 0.7612, 1.9912,
0.1372, 0.1044, 0.3293, 0.1284, 0.2008, 0.1559, 2.7816,
1.2681, 3.1460, 2.0539, 2.4725, 0.2513, 3.2840, 0.8329,
0.3651, 0.2275, 3.4083, 1.4226, 0.4930, 0.1855, 1.0963,
0.0806, 0.7762, 0.3913, 4.1721, 2.0489, 0.1123, 0.8264,
1.2338, 1.4394, 2.6017, 0.1304, 0.1941, 0.3893, 0.0812,
0.2379, 0.2127, 0.5499, 1.1194, 0.2605, 1.5793, 3.3719,
0.6777, 0.4231, 0.5544, 0.1449, 4.3974, 0.1616, 0.1133,
0.9677, 4.5582, 2.2352, 1.8462, 3.3296, 0.6002, 2.0930,
2.5426, 2.0889, 0.1057, 1.0226, 2.0386, 0.1605, 1.5913,
1.4888, 0.1281, 0.5203, 1.2742, 0.0931, 0.1346, 0.2276,
0.5288, 0.5140, 0.8369, 0.2052, 1.5964, 1.9477, 1.3792,
0.0607, 0.9006, 0.1008, 1.0014, 0.1392, 0.1367, 0.2157,
1.3167, 0.0771, 0.6055, 1.3617, 0.0830, 0.7960, 0.2087,
1.0603, 3.0961, 1.4447, 0.0836, 1.1496, 1.8104, 1.4081,
1.7148, 0.6299, 0.2423, 1.0401, 1.4651, 1.9514, 0.1030,
0.5741, 0.9484, 0.1857, 2.3453, 0.1309, 1.4680, 0.1312,
0.8779, 0.2468, 0.1915, 0.4950, 0.2295, 2.2263, 0.1242,
0.9972, 0.3473, 0.0981, 2.1456, 0.3255, 0.6491, 2.3015,
0.1871, 2.1784, 0.1374, 1.4600, 1.4225, 0.3854, 0.5185,
1.4977, 2.2043, 0.8994, 0.7501, 0.2313, 0.1491, 1.2269,
1.2909, 1.3430, 1.1679, 1.5179, 1.8060, 2.0464, 0.4479,
0.0582, 0.7487, 0.4906, 0.0855, 0.3687, 1.1494, 2.8291,
0.2359, 0.6281, 0.1782, 0.1051, 0.1109, 0.8835, 2.0095,
0.2279, 0.0574, 0.5812, 0.1419, 0.4785, 0.1709, 0.3473,
0.1116, 3.0928, 0.8682, 1.8383, 1.6430, 0.1077, 0.5737,
0.3237, 0.3110, 0.3573, 1.6076], device='cuda:0')), ('layer2.0.conv1.weight', tensor([[[[ 1.0023e-02]],
[[ 1.5478e-02]],
[[ 2.5978e-02]],
...,
[[ 1.2872e-02]],
[[-3.1680e-02]],
[[ 5.8940e-03]]],
[[[ 6.2445e-03]],
[[ 7.5096e-03]],
[[-2.5023e-02]],
...,
[[ 7.0060e-03]],
[[-1.8096e-02]],
[[-6.7320e-02]]],
[[[ 2.6610e-02]],
[[ 1.8790e-03]],
[[-6.9333e-02]],
...,
[[ 9.8718e-03]],
[[-1.8730e-02]],
[[-3.0576e-03]]],
...,
[[[-3.4748e-03]],
[[ 3.9990e-03]],
[[-8.3690e-04]],
...,
[[ 8.5643e-03]],
[[ 1.6422e-02]],
[[ 4.1199e-03]]],
[[[ 4.9666e-04]],
[[ 2.2655e-03]],
[[ 1.2856e-01]],
...,
[[-3.1404e-03]],
[[-7.7637e-03]],
[[ 4.9525e-02]]],
[[[ 1.0553e-02]],
[[ 8.9444e-04]],
[[-1.3065e-02]],
...,
[[ 5.8324e-03]],
[[ 2.9158e-02]],
[[-4.8332e-02]]]], device='cuda:0')), ('layer2.0.bn1.weight', tensor([ 0.2639, 0.1874, 0.2028, 0.2016, 0.2028, 0.2079, 0.2166,
0.1769, 0.1823, 0.1636, 0.2489, 0.1522, 0.1968, 0.2547,
0.1898, 0.2004, 0.2001, 0.1535, 0.2109, 0.1407, 0.2053,
0.1700, 0.2409, 0.1672, 0.2550, 0.1277, 0.2347, 0.1519,
0.3125, 0.2502, 0.1831, 0.1560, 0.1685, 0.1832, 0.2342,
0.1888, 0.1630, 0.2165, 0.3512, 0.1941, 0.2384, 0.1560,
0.2359, 0.1366, 0.1788, 0.2182, 0.1845, 0.1800, 0.1408,
0.2376, 0.2027, 0.2410, 0.2098, 0.2116, 0.1076, 0.1869,
0.2270, 0.2286, 0.1911, 0.1916, 0.2162, 0.2079, 0.2235,
0.2123, 0.2124, 0.2164, 0.2602, 0.1321, 0.1902, 0.1735,
0.2416, 0.1990, 0.2038, 0.2215, 0.1764, 0.2177, 0.2127,
0.2317, 0.2532, 0.1651, 0.1626, 0.1963, 0.2081, 0.1926,
0.2130, 0.2600, 0.1959, 0.1637, 0.2128, 0.3089, 0.1876,
0.1892, 0.1551, 0.2224, 0.1925, 0.2220, 0.2031, 0.2480,
0.2083, 0.2479, 0.1450, 0.2119, 0.1960, 0.2159, 0.2165,
0.1843, 0.2348, 0.1779, 0.2566, 0.2222, 0.2028, 0.1639,
0.2372, 0.1999, 0.2228, 0.2090, 0.2407, 0.2054, 0.2224,
0.2153, 0.2855, 0.2180, 0.1861, 0.1533, 0.2073, 0.2611,
0.1852, 0.2868], device='cuda:0')), ('layer2.0.bn1.bias', tensor([-0.1599, -0.0919, -0.1868, -0.0968, -0.0293, -0.1193, -0.0766,
-0.1058, -0.0625, 0.0841, -0.1555, 0.0708, -0.2071, -0.1594,
-0.1094, -0.0251, -0.0778, 0.0911, -0.0694, 0.0664, -0.0250,
0.0363, -0.1849, 0.0753, -0.1483, 0.1161, -0.1218, 0.1143,
-0.1213, -0.1672, 0.0653, 0.0205, -0.0308, 0.0154, -0.1299,
0.0465, 0.0827, -0.0653, -0.1969, 0.0921, -0.0810, 0.0852,
-0.1423, 0.0951, 0.0946, -0.1390, 0.0414, 0.0242, 0.0930,
-0.1663, -0.0784, -0.0960, -0.1124, -0.1654, -0.1033, -0.0265,
-0.0821, -0.0763, -0.0890, -0.0049, -0.0760, -0.1013, -0.1025,
-0.1701, -0.1045, -0.0820, -0.2578, 0.0469, -0.1343, 0.0628,
-0.1718, 0.0618, -0.0345, -0.0653, 0.0661, -0.0572, -0.0401,
-0.1043, -0.1747, 0.0528, 0.0223, -0.0988, -0.1399, -0.0677,
-0.0476, -0.1639, -0.1153, 0.0008, -0.1069, -0.2375, -0.0232,
-0.0219, -0.1606, 0.0245, 0.0211, -0.1168, -0.1102, -0.1294,
-0.0866, -0.2173, 0.0296, -0.0928, -0.0338, -0.0180, -0.1379,
0.0014, -0.0956, 0.1095, -0.2016, -0.1424, -0.1554, 0.0160,
-0.1674, -0.0748, -0.1505, -0.0514, -0.1245, -0.1391, -0.0912,
-0.1194, -0.3035, -0.2166, -0.1192, 0.0447, -0.1237, -0.1406,
0.0175, -0.2559], device='cuda:0')), ('layer2.0.bn1.running_mean', tensor([-0.0781, -0.0289, -0.1624, -0.0476, 0.1549, -0.0135, -0.2048,
-0.2088, 0.0365, 0.1477, -0.0510, -0.0846, 0.0356, -0.0315,
-0.2131, -0.1171, -0.1629, -0.0860, -0.1558, 0.0478, -0.2467,
-0.0226, 0.0832, 0.0896, 0.3585, -0.2514, 0.0657, 0.0094,
-0.0263, 0.1040, 0.1466, -0.1878, -0.0839, 0.1170, -0.1187,
-0.1226, -0.0350, -0.0670, -0.0523, 0.0749, -0.1505, -0.0118,
-0.1860, -0.0105, 0.1370, -0.1270, -0.0176, -0.0388, -0.0289,
0.1434, -0.1645, 0.0336, -0.3019, 0.0873, 0.1461, -0.1040,
-0.3803, -0.0418, -0.1395, -0.1907, -0.0613, -0.1144, 0.0728,
-0.2925, -0.1372, -0.1711, -0.2763, -0.0221, -0.0675, 0.0049,
-0.0796, 0.0896, 0.0392, -0.1416, 0.0789, -0.0793, -0.2526,
-0.1337, 0.1387, -0.0682, 0.0149, -0.3041, -0.0847, -0.1940,
0.0150, 0.0403, -0.1738, -0.0636, -0.2580, -0.0249, -0.1961,
0.0364, 0.1176, 0.0824, -0.0700, -0.1521, -0.2376, -0.0980,
0.0089, -0.3064, -0.0411, -0.0927, -0.1859, -0.1848, -0.4024,
0.1665, -0.0981, 0.0265, -0.2118, -0.1941, -0.0840, -0.0338,
-0.0177, -0.3567, -0.2194, -0.1358, -0.1312, -0.1744, -0.0787,
-0.0775, -0.3283, -0.1888, -0.1161, 0.0521, -0.2345, -0.0623,
0.0123, -0.0271], device='cuda:0')), ('layer2.0.bn1.running_var', tensor([ 0.0373, 0.0168, 0.0146, 0.0210, 0.0335, 0.0136, 0.0263,
0.0100, 0.0164, 0.0237, 0.0257, 0.0151, 0.0066, 0.0269,
0.0155, 0.0319, 0.0196, 0.0216, 0.0305, 0.0139, 0.0354,
0.0186, 0.0229, 0.0252, 0.0257, 0.0203, 0.0291, 0.0264,
0.1838, 0.0190, 0.0251, 0.0186, 0.0215, 0.0364, 0.0193,
0.0314, 0.0262, 0.0368, 0.1665, 0.0446, 0.0390, 0.0279,
0.0286, 0.0150, 0.0362, 0.0139, 0.0179, 0.0216, 0.0169,
0.0222, 0.0200, 0.0292, 0.0169, 0.0077, 0.0037, 0.0300,
0.0332, 0.0467, 0.0312, 0.0447, 0.0356, 0.0193, 0.0256,
0.0145, 0.0299, 0.0298, 0.0236, 0.0126, 0.0080, 0.0221,
0.0228, 0.0449, 0.0361, 0.0291, 0.0250, 0.0281, 0.0212,
0.0221, 0.0331, 0.0166, 0.0214, 0.0259, 0.0215, 0.0262,
0.0232, 0.0291, 0.0216, 0.0157, 0.0278, 0.0152, 0.0143,
0.0326, 0.0061, 0.0452, 0.0416, 0.0328, 0.0243, 0.0233,
0.0170, 0.0187, 0.0209, 0.0195, 0.0198, 0.0495, 0.0256,
0.0350, 0.0428, 0.0339, 0.0220, 0.0168, 0.0124, 0.0281,
0.0219, 0.0184, 0.0261, 0.0341, 0.0423, 0.0171, 0.0255,
0.0240, 0.0259, 0.0158, 0.0101, 0.0137, 0.0307, 0.0325,
0.0305, 0.0307], device='cuda:0')), ('layer2.0.conv2.weight', tensor([[[[ 5.7148e-03, 5.4153e-03, 6.9670e-03],
[-1.4907e-03, 8.5546e-03, 1.2571e-02],
[ 5.9339e-03, 1.9393e-02, 2.3238e-02]],
[[-5.9089e-03, -2.7058e-02, -2.0096e-02],
[ 8.4784e-03, 1.2187e-02, 1.0419e-02],
[ 1.4465e-02, 2.3546e-02, 3.6228e-02]],
[[ 6.5031e-03, -1.2364e-02, 3.6871e-03],
[-4.0839e-03, -7.5281e-03, -3.6343e-03],
[ 1.2317e-02, 4.9689e-03, 4.7140e-03]],
...,
[[-8.9779e-03, -9.8717e-03, -1.4437e-02],
[-9.0001e-03, 1.9672e-03, -3.5889e-03],
[-3.8357e-03, -1.4406e-02, 1.8059e-02]],
[[-4.6118e-02, -2.8350e-02, -1.6736e-02],
[-2.8602e-02, -4.6270e-03, -1.8925e-02],
[-2.6982e-02, 2.7646e-04, 2.4983e-02]],
[[-8.5529e-04, -1.3946e-03, -1.4319e-02],
[ 1.1076e-02, 1.8380e-03, -1.5675e-02],
[-5.8583e-03, -2.3064e-02, -1.7122e-02]]],
[[[ 3.1264e-02, 1.4076e-02, 1.3915e-02],
[ 1.7699e-02, 2.3030e-02, 3.1258e-02],
[ 1.2741e-02, 1.8671e-02, 6.9519e-03]],
[[ 2.0218e-03, 9.5968e-03, -9.3861e-03],
[-1.0374e-02, -9.1659e-04, -6.3050e-03],
[ 4.4324e-03, 6.6241e-03, -1.7597e-02]],
[[-6.2299e-03, -2.0637e-02, 2.5680e-02],
[-1.3865e-02, -2.1147e-03, 1.1478e-02],
[-2.1541e-02, -6.0607e-03, 1.5434e-02]],
...,
[[-1.4400e-03, -2.9872e-02, -2.6718e-02],
[-2.5183e-02, -2.0159e-02, -2.6662e-02],
[-2.2512e-02, -2.2985e-02, -1.4249e-02]],
[[ 2.6493e-02, 3.5156e-02, 2.7891e-02],
[ 1.8545e-02, -3.7416e-03, 2.5659e-02],
[ 3.0646e-02, 4.1930e-02, 3.6475e-02]],
[[ 1.8742e-02, 2.4336e-02, 8.5196e-04],
[ 2.9210e-02, 3.9976e-02, 2.5880e-02],
[ 1.7010e-02, 3.6066e-02, 1.3382e-02]]],
[[[ 1.2474e-02, -1.8047e-02, -4.6260e-02],
[ 2.9568e-02, 3.4341e-03, -2.9960e-02],
[ 1.6432e-02, -8.8087e-03, -2.7499e-02]],
[[-8.4319e-03, -9.7227e-03, 1.6514e-02],
[-1.3975e-03, 4.4135e-03, 1.8744e-03],
[ 3.7071e-04, -9.3388e-03, 1.8634e-03]],
[[-2.6396e-02, 1.1149e-02, 3.6830e-02],
[-2.0837e-02, 4.7213e-03, 3.5836e-02],
[-3.2991e-02, -1.6638e-02, 1.7730e-02]],
...,
[[-1.5793e-02, -2.7924e-03, 1.0733e-02],
[ 1.4940e-02, -1.6132e-03, 1.6853e-03],
[ 1.0718e-03, -1.7483e-02, 3.1942e-03]],
[[ 4.7515e-02, 4.0208e-02, 1.0953e-02],
[ 2.3981e-02, 3.2340e-03, 7.9321e-04],
[ 1.2609e-02, 3.7900e-02, -1.8043e-03]],
[[-1.1543e-02, -1.8122e-02, -3.7408e-02],
[-2.3050e-03, -1.0188e-02, -4.5527e-02],
[-1.3501e-02, -2.7866e-02, -4.3160e-02]]],
...,
[[[-8.0249e-03, -2.4573e-02, -2.3936e-02],
[-2.5406e-02, -1.1022e-02, -4.0906e-02],
[-1.5074e-02, -2.1930e-02, -4.7509e-02]],
[[-3.8590e-03, -8.7336e-04, -7.7881e-03],
[-4.3808e-03, 1.8722e-03, -1.3124e-02],
[-2.7823e-03, -1.1781e-02, -1.0455e-02]],
[[-1.8092e-02, -1.3784e-02, 1.5378e-02],
[-2.8538e-02, -1.7802e-02, 2.6612e-02],
[ 1.4466e-02, 1.9868e-02, 6.0853e-02]],
...,
[[-1.3183e-02, 3.3895e-04, 4.1337e-03],
[ 1.2881e-03, -1.0550e-02, 1.2804e-04],
[ 3.4866e-03, -1.1056e-02, -1.6554e-02]],
[[-1.3392e-02, 2.8013e-02, 1.2245e-02],
[-1.9193e-02, 9.0613e-03, 3.3513e-03],
[-3.5666e-02, -1.0572e-02, -1.8500e-02]],
[[-6.6079e-03, -1.2106e-03, -2.2793e-02],
[-9.2909e-03, -7.9433e-03, -1.9724e-02],
[ 1.5923e-03, -9.6041e-04, -1.7361e-02]]],
[[[ 1.1845e-02, 8.5594e-03, -1.7909e-02],
[ 8.7121e-03, -3.6880e-04, 1.4275e-03],
[-1.4048e-02, -8.8601e-05, 1.3824e-02]],
[[-4.2661e-03, -9.6938e-03, 3.0361e-02],
[-1.6280e-02, -1.1626e-03, 1.6717e-02],
[-1.3279e-02, 3.4856e-03, -1.8792e-02]],
[[-4.2527e-03, 1.2455e-02, -1.4695e-03],
[ 4.5873e-03, 3.0026e-03, -4.5780e-03],
[ 8.0806e-03, 5.6422e-03, -5.4317e-03]],
...,
[[ 2.3770e-03, -8.0848e-04, 1.3950e-02],
[ 6.4999e-03, 8.1204e-04, -8.5034e-03],
[-5.5089e-03, 2.6967e-02, -2.2525e-02]],
[[ 3.9749e-02, -6.4180e-04, -2.3314e-02],
[ 1.1120e-02, -1.9094e-03, -2.6408e-02],
[ 2.1081e-02, -2.3960e-02, 1.6082e-02]],
[[-1.9112e-02, -2.0600e-02, -8.8751e-03],
[-2.0385e-02, -2.6840e-02, 6.0018e-03],
[-1.9251e-02, -2.0094e-02, -3.1182e-03]]],
[[[-1.3227e-02, -2.3923e-02, 4.6463e-03],
[-1.7025e-02, -2.8171e-02, -1.7361e-02],
[-3.2787e-02, -4.1742e-02, -2.6234e-02]],
[[ 1.5639e-02, 2.0191e-02, 1.4607e-02],
[ 3.0779e-02, 1.9010e-02, 2.9254e-02],
[-9.7529e-03, 1.3953e-02, 8.8613e-03]],
[[-3.7030e-03, -2.7738e-03, -2.1177e-03],
[ 1.5293e-02, -3.9039e-03, -1.7507e-02],
[ 6.5572e-03, 3.6534e-02, 2.9290e-02]],
...,
[[ 2.8785e-02, -6.0905e-03, -1.0742e-02],
[ 1.4855e-02, 2.5195e-02, -5.5631e-03],
[-2.0345e-02, -3.4998e-03, -8.5412e-03]],
[[-4.8109e-03, 1.5368e-02, 2.5238e-02],
[-2.4375e-02, 1.7427e-02, -2.7583e-03],
[ 9.9079e-03, -3.3578e-02, -3.2155e-02]],
[[-1.0389e-02, -8.7084e-03, -7.3160e-03],
[ 1.3486e-02, -4.3148e-03, -8.4715e-03],
[ 5.3262e-03, -5.1335e-03, -5.3968e-03]]]], device='cuda:0')), ('layer2.0.bn2.weight', tensor([ 0.2249, 0.1925, 0.1457, 0.2466, 0.2248, 0.2264, 0.1834,
0.1699, 0.1916, 0.2087, 0.2821, 0.1886, 0.1893, 0.2142,
0.1779, 0.1907, 0.1472, 0.2304, 0.2075, 0.2326, 0.2907,
0.2000, 0.1991, 0.1983, 0.1849, 0.1865, 0.1913, 0.2084,
0.2242, 0.1907, 0.2041, 0.1485, 0.2168, 0.2858, 0.2051,
0.2691, 0.2394, 0.2259, 0.1847, 0.1548, 0.2462, 0.2430,
0.2605, 0.1886, 0.1899, 0.1853, 0.1896, 0.2061, 0.1745,
0.2505, 0.1547, 0.1989, 0.1629, 0.2077, 0.2306, 0.1965,
0.1766, 0.2596, 0.2209, 0.1635, 0.2007, 0.2241, 0.1531,
0.2027, 0.1780, 0.1609, 0.2262, 0.1849, 0.1500, 0.2041,
0.1681, 0.1448, 0.2372, 0.2724, 0.2123, 0.1837, 0.1885,
0.2116, 0.1979, 0.1482, 0.2075, 0.1950, 0.1618, 0.2237,
0.2049, 0.1529, 0.2331, 0.2129, 0.2181, 0.1866, 0.2668,
0.1958, 0.2223, 0.1585, 0.1602, 0.1935, 0.2472, 0.2191,
0.1996, 0.1802, 0.2226, 0.2441, 0.1486, 0.1764, 0.1764,
0.1997, 0.2210, 0.2333, 0.1666, 0.2179, 0.1985, 0.1715,
0.1806, 0.2031, 0.2108, 0.2498, 0.1628, 0.2023, 0.1595,
0.2039, 0.2070, 0.2019, 0.2314, 0.1813, 0.1791, 0.1618,
0.2178, 0.1765], device='cuda:0')), ('layer2.0.bn2.bias', tensor([-0.0598, 0.1500, 0.1551, -0.0595, -0.0690, -0.1054, 0.0655,
0.0134, -0.0076, -0.0359, -0.0323, 0.0047, -0.0527, 0.0220,
-0.0195, 0.1606, 0.0690, -0.0332, -0.0221, -0.0216, -0.0347,
0.0795, -0.0550, 0.0044, 0.0412, -0.0321, -0.0166, -0.0164,
-0.0345, -0.0317, -0.0186, 0.1326, -0.0612, -0.0387, 0.0007,
-0.1574, 0.0054, -0.0069, -0.0222, 0.1729, -0.0216, -0.0909,
-0.1080, -0.0130, 0.0267, 0.1133, -0.0435, -0.0617, 0.1678,
-0.0375, 0.1510, -0.0130, 0.2467, -0.0718, -0.0950, 0.0546,
0.1932, -0.0419, 0.0211, 0.0493, -0.0233, -0.0336, 0.2257,
0.0171, 0.0563, 0.0568, -0.0530, -0.0064, 0.2224, -0.0075,
0.2084, 0.1459, -0.0542, -0.1133, -0.0431, 0.0079, 0.0875,
-0.0350, -0.0037, 0.2025, -0.0222, -0.0094, 0.1412, -0.0589,
-0.0062, 0.1147, -0.0543, -0.0114, -0.0473, 0.0007, -0.3161,
0.0604, 0.0197, 0.0256, 0.0456, 0.1723, -0.0516, 0.0427,
0.0975, -0.0074, -0.0679, -0.1042, 0.2585, -0.0494, 0.1126,
0.0128, -0.0250, -0.0299, 0.2048, -0.0451, -0.0578, 0.0532,
0.0499, 0.0046, -0.0285, -0.1081, 0.1042, 0.0073, 0.1768,
0.0838, 0.0246, 0.1291, -0.0364, 0.0228, 0.0882, 0.1532,
-0.0570, 0.0508], device='cuda:0')), ('layer2.0.bn2.running_mean', tensor([-0.1384, -0.1698, 0.0093, -0.1065, -0.0636, 0.0466, -0.0052,
-0.0921, -0.1068, -0.0388, -0.2716, -0.0473, -0.0749, -0.0381,
-0.2133, -0.1624, -0.1283, -0.1114, 0.0124, 0.0097, -0.1946,
-0.0196, -0.0419, -0.1609, -0.0358, 0.0179, -0.0327, -0.0211,
-0.0727, -0.1240, -0.0063, 0.0008, -0.0511, -0.2189, -0.0721,
-0.2698, -0.0027, -0.1189, -0.0512, -0.0900, -0.0878, -0.1427,
-0.0866, 0.4502, -0.0787, 0.0490, -0.0171, -0.0896, -0.0522,
-0.1662, -0.0046, -0.0397, 0.0443, -0.2211, -0.1814, -0.0437,
0.1307, -0.1561, 0.0079, -0.1363, -0.1119, -0.1325, 0.0375,
-0.0152, -0.0528, -0.0398, 0.0810, 0.0006, -0.0315, -0.1099,
0.1079, 0.0391, -0.0699, -0.0440, -0.0026, -0.0192, -0.0095,
-0.0606, -0.0462, 0.0176, -0.0617, 0.0147, -0.0366, -0.1308,
-0.0761, 0.0369, -0.1305, 0.0206, -0.0631, 0.0656, -0.2360,
-0.0220, -0.0737, -0.0593, -0.0011, -0.1066, 0.0060, -0.0160,
0.0359, 0.2074, -0.0684, -0.1137, -0.0078, -0.0843, -0.0649,
-0.0110, -0.0666, -0.0504, 0.1075, -0.0438, -0.0420, -0.0374,
-0.0506, -0.1934, -0.1703, -0.2015, -0.0753, -0.0874, 0.0359,
-0.0192, -0.0037, 0.0614, -0.1412, -0.0813, -0.0048, -0.0466,
0.0238, -0.0500], device='cuda:0')), ('layer2.0.bn2.running_var', tensor(1.00000e-02 *
[ 2.6818, 6.1464, 1.8368, 2.8838, 3.0074, 2.1492, 3.7271,
1.9795, 2.1896, 2.1662, 5.6332, 2.4964, 1.4424, 2.7853,
1.8091, 7.9828, 1.3578, 2.9915, 2.0401, 3.7348, 2.4422,
2.3394, 2.0024, 3.4702, 2.9725, 1.8081, 2.4319, 1.9755,
2.8897, 2.1280, 2.2416, 3.0565, 2.6649, 5.7000, 2.6910,
2.4987, 3.6173, 3.0183, 2.0197, 3.1076, 7.1252, 2.7301,
2.7651, 1.9119, 3.0144, 3.9931, 2.0541, 1.9363, 2.9769,
6.2772, 1.7601, 2.1551, 3.1537, 2.8651, 2.3136, 3.3620,
3.2976, 3.4972, 3.6715, 2.3695, 1.8913, 3.5567, 2.4717,
2.3482, 3.4518, 1.4884, 2.5542, 1.9754, 2.4605, 2.0859,
3.5534, 2.4395, 3.2819, 2.1078, 2.0506, 2.1313, 3.1246,
1.8070, 2.2204, 2.6746, 1.9324, 2.1715, 3.3050, 2.7764,
2.7183, 2.5421, 3.2537, 2.2937, 2.5260, 2.4147, 1.1274,
4.2444, 3.0421, 1.8865, 1.7334, 6.7181, 3.6373, 3.2582,
3.0125, 2.6954, 2.0425, 3.1739, 2.6017, 1.7399, 2.8741,
2.4839, 1.8959, 3.5319, 2.8759, 1.6634, 1.9163, 2.4310,
2.0252, 2.6119, 1.8112, 3.4105, 1.6091, 2.1799, 2.3995,
3.3236, 2.9057, 3.6157, 2.9658, 2.2053, 2.7017, 2.0977,
2.3119, 2.9194], device='cuda:0')), ('layer2.0.conv3.weight', tensor([[[[ 7.6208e-03]],
[[-6.1014e-03]],
[[ 1.3127e-02]],
...,
[[ 1.0091e-02]],
[[-1.3378e-02]],
[[-6.2687e-03]]],
[[[-2.7508e-02]],
[[-4.3027e-03]],
[[ 2.5836e-02]],
...,
[[ 2.7279e-03]],
[[-9.3313e-03]],
[[ 6.1833e-02]]],
[[[-1.7879e-03]],
[[-9.0075e-04]],
[[-1.7440e-04]],
...,
[[-1.9249e-04]],
[[-3.8543e-06]],
[[-3.3952e-04]]],
...,
[[[ 1.1697e-07]],
[[ 3.4833e-07]],
[[-1.7422e-07]],
...,
[[-1.8790e-07]],
[[-8.5461e-08]],
[[ 2.0189e-07]]],
[[[ 4.2819e-02]],
[[-1.2952e-01]],
[[-5.2879e-02]],
...,
[[ 1.2145e-01]],
[[ 4.3815e-02]],
[[-1.2587e-02]]],
[[[ 5.9202e-04]],
[[-1.4324e-02]],
[[ 7.7022e-03]],
...,
[[ 1.7542e-02]],
[[ 6.5371e-03]],
[[ 8.5100e-03]]]], device='cuda:0')), ('layer2.0.bn3.weight', tensor([ 3.8543e-03, 2.0187e-01, -1.0649e-03, 1.3887e-01, 6.4771e-03,
-2.0702e-06, 2.0280e-01, 1.1607e-02, -4.1076e-03, 2.7771e-02,
8.9216e-02, 1.0155e-01, 1.0902e-01, 3.9999e-02, 4.5550e-06,
1.2153e-01, 2.9691e-01, -1.7842e-02, 1.3224e-01, 9.2594e-02,
8.6137e-02, 2.3588e-03, 2.1034e-01, 1.5372e-07, 1.6275e-01,
-1.0454e-02, 1.1453e-01, 1.5497e-01, 6.6282e-06, 1.6580e-01,
3.1317e-01, -2.4331e-03, 5.2280e-02, 1.5141e-01, 8.9488e-02,
1.9418e-01, 7.4374e-02, 1.9285e-01, 2.3015e-03, 2.2057e-01,
2.4289e-01, 1.8490e-01, 6.0817e-02, 2.4235e-03, -7.2726e-03,
2.1065e-01, 1.1995e-02, 1.9941e-01, 1.5138e-01, 2.0406e-01,
-3.2079e-03, 1.2165e-04, 6.7944e-02, 9.9864e-02, 2.7700e-03,
2.0632e-01, 4.6338e-02, 8.8209e-02, 2.0853e-01, -7.7076e-08,
-1.2983e-03, 1.4655e-01, 1.3156e-01, 6.1504e-03, 1.5142e-01,
1.4627e-01, 6.0212e-02, 1.8846e-01, -1.0229e-06, 2.2133e-01,
2.3804e-01, 1.2794e-01, 2.4388e-01, 1.4702e-01, 3.1081e-01,
9.2273e-03, 2.3980e-01, -8.6978e-03, 1.7617e-01, 1.8940e-01,
2.2123e-01, 1.2454e-01, -1.1824e-02, 2.1235e-01, 1.0170e-01,
1.9985e-01, 1.7297e-01, 9.3104e-04, -5.2291e-04, 2.3814e-06,
6.4092e-07, 2.2321e-01, 1.6644e-01, 1.3561e-02, 2.6848e-08,
1.7824e-01, 1.9010e-01, -3.5674e-02, 1.6651e-01, -9.1091e-04,
1.2540e-01, 1.7982e-01, 9.7606e-03, 1.9054e-01, 4.5191e-03,
-1.9165e-04, 1.6682e-01, 1.8502e-01, 1.4833e-01, 4.4347e-02,
1.1528e-01, 1.4835e-01, -1.0824e-07, 1.6343e-01, 1.0696e-02,
1.2262e-01, 1.8079e-01, 1.5248e-01, 1.8994e-01, 6.9889e-02,
1.0647e-01, 1.2310e-01, 6.5127e-02, 1.1183e-01, 1.0039e-01,
-2.6980e-03, 2.3881e-01, 5.7536e-02, 6.6677e-02, 1.5345e-01,
1.6415e-01, 1.4292e-01, 1.0783e-01, 1.7668e-06, 9.5887e-02,
7.9922e-02, 1.4580e-01, 1.8263e-01, 4.5771e-07, 1.7064e-01,
1.1783e-01, 1.3707e-01, 2.1131e-01, 1.6595e-04, -3.2211e-02,
-1.5293e-02, -1.1494e-03, 1.8486e-01, 2.6609e-01, 1.8440e-01,
2.2997e-01, 4.7652e-02, 9.2821e-02, -5.9602e-06, 5.3773e-03,
1.2814e-01, -6.3296e-04, 2.2609e-06, 1.6365e-01, 1.1699e-01,
1.9503e-01, 1.8941e-03, 1.7983e-01, 1.1410e-01, 1.8587e-01,
2.6753e-04, 1.1377e-01, 1.4556e-01, 2.0240e-01, 1.0300e-01,
2.5308e-08, 2.0455e-01, 2.3635e-01, 4.0376e-03, 1.4885e-06,
2.6631e-01, 7.2618e-02, 2.5890e-01, -4.5141e-03, 1.4874e-01,
2.5813e-01, 1.4446e-01, 2.0476e-01, 5.1917e-04, 1.3332e-01,
-3.0500e-03, -2.8465e-07, 1.1479e-01, 5.7585e-03, 1.0129e-01,
2.1036e-01, 1.7736e-01, 1.0850e-01, 7.3646e-02, -8.7343e-04,
1.8930e-01, 2.5171e-08, 2.2083e-01, 3.7008e-02, -8.7625e-03,
2.6809e-06, 1.5259e-01, 2.4068e-01, -4.6558e-03, -2.5157e-08,
1.7323e-01, 1.4529e-01, 1.5122e-01, 1.3936e-02, 1.0643e-02,
1.0234e-01, 2.1276e-01, 1.1609e-01, 1.4455e-06, 7.7813e-02,
2.3590e-01, 1.0331e-01, -2.1370e-03, 2.2225e-02, 2.6490e-01,
5.3318e-02, 1.1014e-01, -4.7809e-09, 1.5605e-01, 1.3709e-01,
1.9997e-01, 1.2544e-01, 1.6009e-03, 1.7971e-01, 2.3313e-01,
1.6170e-01, 7.9911e-02, -3.7592e-03, 5.8648e-02, -5.9956e-09,
1.7753e-01, 3.2006e-06, 2.0509e-01, 2.4585e-01, 1.8359e-01,
-9.5735e-06, 6.1166e-03, 1.9506e-01, -1.4920e-02, 1.2797e-01,
1.5274e-01, 1.1678e-01, 2.5840e-01, 1.8168e-01, -3.5162e-03,
1.7881e-01, 2.2759e-03, 1.7817e-01, 1.5301e-01, 1.2492e-01,
1.6297e-03, 2.2141e-01, 1.5610e-01, 1.7923e-01, 1.9679e-01,
1.1674e-01, 2.5461e-01, 2.2336e-01, -8.6711e-03, 2.1022e-01,
2.6223e-02, 5.8941e-02, 6.7842e-03, 3.9383e-04, -4.8280e-03,
-8.4383e-07, 1.5757e-01, -3.3189e-07, 9.4600e-02, 2.0100e-01,
-4.7420e-03, 8.0226e-02, 4.1372e-03, 1.7892e-01, 2.0364e-01,
1.6549e-01, 9.2902e-03, 1.9418e-01, 2.2028e-03, 1.3965e-03,
2.1172e-03, 2.0174e-01, 1.9476e-01, 2.1752e-01, 1.6050e-01,
1.6618e-02, 1.3520e-04, 2.1298e-01, 2.4628e-01, 2.0706e-01,
2.3657e-01, 1.8144e-01, 8.8419e-06, -1.8546e-02, 2.9771e-04,
1.6264e-01, 1.6636e-01, 1.1508e-01, 1.1071e-01, 1.4062e-01,
1.2329e-01, 1.0167e-01, 9.0378e-02, 6.5264e-02, -1.0424e-02,
1.9850e-01, -1.4600e-06, 4.6740e-08, 6.2084e-07, -4.7988e-03,
6.1632e-02, 1.6973e-01, 1.7514e-03, 7.9060e-08, -4.5426e-03,
1.7435e-01, 1.5014e-01, 1.0646e-01, 9.1071e-02, 2.1939e-01,
2.5915e-08, 1.6844e-01, 1.2294e-01, 1.1205e-01, 8.2247e-02,
1.7394e-01, 2.4874e-01, 1.5983e-07, 2.4351e-03, 1.0579e-01,
7.5377e-02, 2.2991e-01, -9.1592e-03, 2.0529e-01, 7.1128e-02,
9.9701e-04, 1.5408e-01, 2.2712e-02, 2.4965e-01, 1.3048e-03,
1.8778e-01, 8.0938e-08, 1.8622e-02, 1.2455e-01, 1.6796e-02,
1.2836e-01, 1.3399e-02, 2.4071e-01, -5.7628e-03, -8.2883e-04,
1.5164e-01, 1.2500e-01, -1.3223e-03, 1.6686e-01, 2.0293e-01,
1.2653e-01, 1.1740e-01, 1.2208e-01, -5.8454e-04, 1.1730e-05,
2.4734e-01, 9.2554e-02, 2.4105e-01, 1.6128e-01, 1.4836e-01,
1.4868e-07, 6.1657e-03, 1.6485e-01, 1.7401e-01, 2.5618e-01,
2.8015e-01, 1.1606e-01, 1.7603e-01, 1.0104e-01, 9.7931e-03,
3.0543e-01, 1.7581e-01, -3.1169e-03, 1.1331e-01, 2.3405e-01,
1.5033e-01, 1.4255e-06, -2.4860e-03, 1.0574e-01, 2.1443e-01,
2.1830e-01, 2.2319e-01, -3.5428e-03, 7.6133e-02, 2.1730e-01,
-1.3519e-03, 7.5629e-03, 1.1418e-01, 1.4467e-01, 5.1706e-02,
-2.0379e-07, -4.1459e-03, 2.7953e-01, 9.6022e-02, 2.5677e-01,
2.0866e-01, 4.1675e-02, 8.8582e-02, 1.5234e-01, -3.0899e-10,
1.8085e-01, 2.0183e-01, -2.3279e-03, 6.9877e-08, 8.2608e-02,
2.0111e-01, 3.8781e-02, 2.5544e-01, 4.8373e-03, 1.2000e-01,
1.7985e-01, 3.7402e-02, 1.0675e-01, 1.3236e-01, 9.0898e-02,
9.6931e-02, -4.1332e-07, 4.0971e-02, 1.6959e-01, 9.8011e-02,
1.6147e-06, 8.6560e-02, 1.9620e-01, 1.6818e-06, 3.2809e-01,
1.8875e-01, 2.0212e-01, 2.5101e-01, 1.3554e-01, 2.2584e-01,
1.0739e-01, 6.8136e-04, 6.9796e-02, 5.9475e-03, 2.8626e-02,
1.5608e-01, 6.0822e-07, 2.2908e-01, 2.1893e-01, 6.4698e-05,
1.5747e-01, 2.5414e-07, 1.9498e-08, 1.8155e-01, 1.0366e-03,
1.1361e-01, 1.8329e-01, 1.8751e-01, -3.6502e-04, 1.2180e-01,
1.4766e-01, 1.7146e-01, 4.9706e-02, 1.2225e-01, 1.0283e-02,
2.1014e-01, 1.0958e-01, 4.4404e-08, -1.3192e-02, 1.9448e-01,
1.5193e-01, 3.7440e-03, -4.5107e-08, -8.7403e-04, -2.6012e-03,
1.2968e-01, 3.4562e-02, 3.0098e-02, 1.8282e-01, 2.9391e-01,
5.8525e-02, 1.5908e-01, 2.2260e-02, 1.9951e-01, 2.3200e-01,
1.3207e-01, 5.8267e-06, 1.4268e-01, 6.0896e-03, 8.0552e-02,
1.9510e-01, 1.4327e-01, -2.2665e-02, 2.1358e-01, -3.2695e-03,
1.3043e-01, 2.0119e-01, 7.1658e-02, 9.7573e-09, 1.6613e-01,
1.6489e-02, 1.1414e-01, 1.8905e-01, 6.3744e-02, 1.3969e-01,
2.0824e-02, 5.5400e-03, 8.8336e-02, 1.4921e-01, 1.4837e-07,
1.5383e-01, 9.2749e-03], device='cuda:0')), ('layer2.0.bn3.bias', tensor([ 5.0408e-02, -5.3801e-02, -4.9714e-03, 2.0539e-02, 4.7344e-02,
-7.9189e-06, 9.5915e-03, 8.5763e-02, 3.1474e-02, 7.3589e-02,
3.1398e-03, -3.0815e-02, 4.4212e-03, 9.0432e-02, -1.7023e-05,
-2.1265e-02, -3.0664e-02, 3.1546e-02, 4.3062e-02, -7.6598e-04,
4.0609e-02, 7.0054e-02, -3.2688e-03, -8.6219e-07, 2.9686e-02,
6.5213e-02, 3.7850e-03, -3.1191e-02, -3.1902e-05, 4.1727e-02,
-9.7415e-02, 7.2782e-02, 3.0331e-02, -1.1470e-02, 8.0664e-03,
-2.3092e-02, 9.3858e-02, 4.7561e-02, -4.6917e-03, 7.5247e-02,
-2.9921e-02, 1.5897e-02, -2.7721e-02, 4.7655e-02, 7.5995e-03,
-5.5337e-02, 7.0081e-02, -3.3844e-02, 6.6372e-02, -5.6138e-02,
9.5078e-02, -8.3608e-04, 5.9420e-02, 4.6229e-02, 9.6871e-02,
1.0570e-02, -1.0752e-02, -3.7556e-03, 1.1314e-02, -4.0713e-06,
4.4267e-02, -3.9143e-03, -7.9898e-03, 3.3193e-03, -1.0885e-02,
3.3494e-03, -9.2455e-02, 3.9653e-02, -7.8888e-06, -2.4530e-03,
1.3655e-02, 1.8437e-02, -1.1777e-02, -1.0059e-02, -4.4714e-02,
5.1907e-02, -2.4790e-02, 1.0142e-01, 1.8356e-02, -5.6835e-02,
-1.9470e-04, 1.7989e-02, -2.3642e-02, 2.5737e-02, -2.2398e-03,
4.8911e-02, 5.5182e-02, -1.8045e-03, 8.5885e-02, -1.0144e-05,
-7.6508e-06, -4.7996e-02, -5.4346e-02, 3.8387e-02, -1.5890e-07,
-1.3694e-02, 2.1481e-02, 3.7525e-02, -3.2900e-03, -1.4641e-02,
-5.8719e-03, -3.0566e-02, 9.1605e-02, 2.9773e-02, 1.0682e-01,
-8.9958e-04, 3.5209e-02, -4.2207e-02, -3.2497e-03, -3.2672e-02,
8.3064e-02, 4.8578e-02, -2.8526e-06, -1.8117e-02, 1.0543e-01,
7.2242e-02, -7.6185e-03, 2.2616e-03, 3.7346e-02, 4.6880e-02,
-7.9570e-03, 3.6065e-02, 7.7348e-02, 3.2475e-03, -3.0122e-02,
1.5840e-01, -3.3549e-02, 5.9058e-02, -5.4857e-04, -4.5232e-02,
-2.0058e-02, 1.7915e-02, 1.7743e-02, -7.1322e-06, 1.1004e-01,
-7.1586e-03, -7.0238e-04, 3.5450e-03, -1.4194e-06, -5.0723e-03,
1.0719e-02, -2.6846e-02, -1.5724e-02, -4.8029e-04, -4.1229e-02,
1.0429e-02, 4.7818e-02, -2.1555e-02, 2.1728e-02, -2.0473e-02,
2.5582e-05, 3.2087e-02, 6.4179e-02, -6.2677e-05, 1.1763e-01,
3.7077e-02, 8.7900e-02, -9.1733e-06, 3.5007e-02, -1.5780e-02,
-1.4645e-02, 1.1493e-01, 3.5553e-02, 3.8060e-02, 5.2214e-02,
-1.3884e-03, -1.9273e-02, -3.3458e-02, -3.7974e-02, -1.7768e-02,
-1.3500e-06, 7.1225e-03, 9.3039e-02, 1.4888e-01, -2.7238e-05,
-5.6481e-02, 1.0640e-01, -3.8111e-02, 5.7626e-02, -1.1871e-02,
-3.7452e-02, -8.3534e-03, 1.3115e-02, 5.2240e-02, 1.4609e-02,
1.3691e-01, -2.8239e-06, 5.4831e-03, 7.0706e-02, -3.6691e-02,
-1.7748e-03, -1.1044e-02, 2.9012e-02, 1.1499e-01, 1.5405e-01,
7.9285e-03, -1.1732e-07, -2.6413e-02, 1.2214e-02, 8.3662e-03,
-6.2902e-06, 5.0012e-02, -6.5154e-03, 5.0267e-02, -7.1198e-07,
-4.5093e-02, 2.5603e-02, 3.9757e-02, 4.9278e-02, -9.7368e-03,
-1.0309e-03, 8.0354e-02, 4.9503e-02, -1.0135e-05, -2.0469e-02,
-2.3065e-02, -1.0967e-02, 1.3992e-01, 1.3652e-02, 2.2791e-02,
5.5029e-02, -1.3392e-02, -1.2968e-07, -3.2554e-02, 2.6425e-02,
2.0118e-02, 5.8526e-02, 1.9821e-01, 2.1531e-03, 1.3833e-02,
-1.2165e-02, 7.6226e-02, 1.0087e-01, -2.9128e-02, -1.1165e-07,
3.4191e-02, -2.7389e-05, -6.7306e-03, 5.2081e-02, 5.8233e-03,
-1.8436e-04, 1.9059e-01, 2.4796e-02, 8.9079e-02, -2.8740e-03,
-5.1246e-02, 2.3442e-02, 1.3485e-04, -4.4533e-02, 3.7772e-02,
4.5265e-02, 1.2982e-01, -7.0863e-03, 2.9466e-03, 2.1229e-02,
6.8678e-02, -2.4108e-02, -5.9673e-03, 3.3583e-02, 5.1814e-03,
-3.4058e-02, 2.4528e-02, 3.2980e-03, 8.5338e-02, 4.6658e-03,
1.1495e-01, 2.1275e-02, 2.0298e-02, 6.5352e-02, 4.8504e-02,
-1.2237e-04, -2.2201e-02, -3.2690e-06, 5.6547e-02, 9.2143e-03,
5.3177e-02, 1.3121e-02, 5.8778e-02, -1.0628e-02, -4.6647e-02,
-3.8120e-03, 1.6281e-02, -1.6294e-02, 4.6992e-02, 8.2434e-02,
-3.7641e-02, 9.0587e-03, -2.1060e-03, -2.2506e-02, -4.0034e-02,
-1.7160e-02, -9.8877e-04, -4.3668e-02, -4.3654e-02, -7.9761e-02,
8.0995e-04, 3.3363e-02, -9.5408e-05, 2.4635e-02, -1.2186e-03,
8.8857e-03, -2.5290e-02, 1.5415e-02, 4.1727e-02, -1.4499e-02,
-2.5007e-03, -4.2352e-02, 2.3150e-02, 1.0252e-02, -5.0827e-02,
-2.9843e-02, -1.7975e-05, -2.4256e-07, -2.3200e-06, 8.7960e-02,
-1.9594e-02, -1.4285e-02, 1.3767e-01, -5.4095e-07, 7.2719e-02,
1.5425e-02, 2.2584e-03, 1.6702e-02, -1.8719e-03, 4.0983e-02,
-1.5115e-07, -6.9896e-02, -1.1585e-02, -3.8220e-03, 3.9209e-02,
7.9392e-02, -3.1275e-03, -6.0793e-07, -6.5054e-02, -4.6122e-03,
2.2881e-02, -1.5298e-02, 7.1332e-02, -2.5570e-02, -2.5730e-02,
1.0177e-01, -3.8346e-03, 1.1747e-01, -3.2341e-02, 1.2023e-01,
9.6849e-03, -1.8967e-07, 9.6480e-02, -3.4635e-03, 1.3868e-01,
-7.0695e-04, -9.4783e-02, -4.2773e-03, 3.6939e-02, 1.4341e-01,
8.6191e-03, 3.4905e-02, 1.3142e-01, 2.1686e-02, -3.2155e-02,
2.2250e-02, 3.3670e-02, 1.1841e-03, -4.6307e-02, -6.7543e-05,
-1.9290e-02, -1.0294e-03, -2.2324e-02, 6.6470e-02, -2.0465e-02,
-5.0082e-06, 1.1699e-01, 3.5496e-02, 2.2371e-02, 7.6098e-02,
-3.1274e-02, -2.6378e-02, 4.4431e-03, -1.3276e-02, 5.5820e-02,
-2.9341e-02, -3.4241e-04, 1.2948e-01, 4.8526e-02, 1.2148e-03,
1.5532e-02, -5.9827e-05, 9.0663e-02, 6.6560e-02, -2.1857e-02,
-4.2524e-04, 8.4397e-04, 7.5189e-02, -4.0382e-02, -5.7350e-03,
1.3059e-01, 8.9209e-02, -5.2390e-02, 8.3347e-03, 3.6906e-02,
-1.9198e-06, -2.5754e-02, -1.8060e-02, 1.5707e-02, -7.2462e-02,
-5.3932e-02, 3.3604e-02, 5.3753e-03, 9.2454e-03, -9.6343e-07,
1.1055e-02, 9.3140e-03, 1.2375e-01, -2.4052e-07, 4.7903e-03,
-2.4964e-02, 7.8106e-03, -2.5364e-02, -1.7556e-02, 1.1134e-02,
-1.5845e-02, -4.8787e-02, 1.7304e-03, -1.9681e-02, -1.2093e-03,
-1.8620e-02, -1.9253e-06, 6.8844e-03, -1.2531e-02, -2.6413e-04,
-8.7646e-06, 4.8010e-02, 8.7525e-02, -8.0243e-06, -1.7114e-01,
-1.3433e-02, -4.0059e-02, -5.1796e-02, 6.7273e-03, 7.6271e-03,
1.7412e-02, 7.1451e-02, -6.3535e-02, 1.4390e-01, 5.2225e-02,
-2.6857e-02, -2.3525e-06, 9.4138e-03, -2.4442e-03, -1.6837e-04,
-2.2859e-02, -2.3380e-06, -4.7066e-07, 2.9032e-02, -4.4317e-03,
1.2823e-02, 6.2972e-02, -2.8694e-02, 1.0436e-01, 1.2098e-01,
-6.8384e-03, 6.1803e-02, -2.1682e-02, 3.3520e-02, 1.8442e-01,
2.0452e-02, -2.2529e-02, -3.8024e-07, 4.7503e-02, -1.6849e-02,
-3.5622e-02, 1.3090e-01, -1.0422e-07, 6.6521e-02, 8.8123e-02,
1.2963e-02, 6.9495e-02, 2.2898e-03, -8.1541e-03, -5.9840e-02,
-3.7383e-03, 1.6682e-02, 1.8710e-02, -3.3964e-02, -6.9819e-03,
-8.8785e-03, -1.6550e-05, 3.6695e-02, 7.4770e-02, -3.6443e-03,
-2.7702e-02, -5.7924e-02, 5.3831e-02, 3.8519e-02, 9.7022e-02,
-4.2779e-02, -1.2827e-02, -5.4793e-02, -6.0001e-08, 7.1382e-03,
6.7838e-02, -9.0207e-02, -2.3035e-02, -8.5935e-04, 2.0521e-02,
7.8742e-02, 3.6046e-02, -7.2988e-03, -4.1833e-02, -8.7138e-07,
7.5300e-02, 1.6350e-01], device='cuda:0')), ('layer2.0.bn3.running_mean', tensor([-6.2018e-03, 6.3003e-02, 8.4960e-04, 1.5055e-02, 1.3847e-02,
1.4540e-06, -3.5650e-02, 3.1650e-04, -3.9915e-04, 1.0864e-02,
-1.9694e-02, 1.4713e-02, -3.4286e-02, 1.7799e-02, -5.2056e-06,
-8.0491e-02, -6.9699e-02, 3.3425e-03, -7.3067e-02, -3.0838e-02,
-9.8401e-03, 7.8454e-03, -2.1305e-02, -1.9281e-07, 7.0921e-02,
-5.3149e-03, -2.0414e-02, 1.0377e-01, 1.3545e-05, 4.7361e-02,
1.0374e-04, -1.6532e-02, -8.3285e-03, 1.0171e-01, 1.9603e-02,
-5.0936e-02, -6.3814e-02, -4.9284e-02, 3.9552e-03, 8.9107e-02,
3.5031e-03, 6.2920e-02, -2.9925e-02, -1.5572e-02, -8.8842e-03,
-3.7858e-02, 1.0693e-03, -2.5806e-02, -2.9888e-02, -4.5851e-02,
1.1843e-02, -1.7521e-04, 4.5751e-02, 4.2278e-03, 4.8487e-03,
-1.1653e-01, -2.5203e-02, 2.1704e-02, 4.8103e-02, 7.2640e-07,
1.8617e-04, 3.1514e-02, -2.8315e-02, 3.8213e-03, 7.1215e-02,
1.0123e-02, 1.4743e-02, 1.3083e-02, -1.1851e-06, 4.7172e-02,
4.9369e-02, -3.3853e-02, -5.7314e-02, 8.1011e-02, -2.5958e-02,
-3.5659e-02, 1.0552e-01, 7.2788e-03, -7.0131e-02, 1.2137e-03,
1.0301e-02, 1.6776e-02, 7.6988e-04, -8.8039e-02, -3.2235e-02,
7.9868e-02, -5.2975e-02, 1.5968e-03, -8.1101e-04, 1.8032e-06,
1.5597e-06, 2.2084e-02, -3.6830e-02, 1.7755e-02, 2.2687e-08,
-8.2168e-03, 2.4323e-02, -4.7401e-02, 8.0776e-02, 1.5797e-03,
9.1618e-02, 5.5884e-02, -3.5922e-03, 5.3335e-03, 2.7186e-03,
-3.5292e-04, 5.0208e-02, -5.8018e-02, 6.1569e-02, -2.1488e-02,
8.9749e-03, 5.2449e-02, 4.6264e-07, 3.6005e-03, 4.8168e-03,
7.9242e-02, 8.4946e-02, 1.3370e-02, -5.8866e-02, -4.7788e-02,
-6.0953e-04, -2.3429e-02, 6.4723e-02, -1.2996e-02, -1.2823e-04,
1.3041e-02, -4.3422e-02, 8.7196e-03, -2.5296e-02, 4.5250e-02,
2.9757e-03, 4.2734e-02, -4.0833e-02, -3.9718e-06, -7.3359e-02,
-3.4823e-02, 6.5440e-03, -4.9368e-02, 5.3716e-07, -6.5478e-03,
2.1737e-03, 8.0760e-03, 2.8964e-02, -1.4499e-04, -5.5908e-03,
-2.2478e-02, -3.8169e-05, 1.9402e-02, -2.9473e-01, 1.0913e-01,
1.8252e-02, -3.2029e-02, 2.1140e-03, -1.0222e-05, -1.7318e-03,
-1.8882e-02, -2.3356e-04, -1.0823e-06, 1.2251e-02, -2.8881e-02,
1.4977e-02, -2.2630e-02, 3.7084e-02, 1.0619e-01, 1.6427e-02,
-5.3891e-04, 1.1692e-01, 2.3470e-02, 6.4214e-02, 5.6913e-02,
1.1813e-07, 2.7628e-02, 1.6939e-01, -1.5842e-02, 2.2865e-06,
-3.2875e-02, 7.3604e-02, 6.0113e-03, 4.7041e-03, 7.0228e-03,
-6.9879e-02, -5.9836e-02, -1.7548e-02, -4.5966e-03, 6.2667e-02,
-2.7980e-02, -1.2257e-06, 1.8308e-02, 9.2542e-03, 9.3524e-02,
-1.2593e-02, -1.5655e-02, 4.9079e-02, -1.3671e-01, 1.9110e-02,
2.0828e-02, -4.5415e-08, 3.3391e-02, 8.9793e-03, -1.0573e-02,
5.0298e-07, 3.8596e-02, 1.0121e-01, -1.0602e-02, -3.7259e-07,
-5.2744e-02, 7.0048e-02, 3.3014e-02, -3.2893e-02, -8.1825e-03,
-8.5603e-02, -2.8417e-03, -9.0147e-02, 1.5946e-06, -8.4105e-03,
-3.0795e-03, -5.8761e-02, 3.1994e-02, -9.3032e-03, -9.7456e-02,
2.2880e-02, 1.2741e-02, -5.7413e-08, 4.8681e-03, 4.7204e-02,
3.6414e-02, -7.8480e-03, 2.3236e-02, 3.5941e-02, -2.0214e-01,
8.3714e-02, -3.2455e-02, 5.1382e-03, 7.7567e-03, -5.8397e-08,
1.6392e-02, 1.1912e-05, 1.1323e-02, 4.8460e-02, -1.1910e-02,
1.3171e-05, 9.6246e-03, 4.8091e-02, -3.0223e-02, -1.8090e-03,
-3.8544e-02, 1.0133e-02, -8.7468e-02, 4.0773e-02, -1.0625e-02,
-1.7836e-02, -2.4670e-02, 6.3115e-03, -3.9420e-02, 3.6707e-02,
6.9660e-03, 1.8957e-02, 5.2998e-02, -2.0478e-02, -5.1350e-02,
7.8511e-03, -8.8431e-03, 2.2146e-02, -2.2835e-02, 4.3533e-02,
-1.0164e-02, -3.4600e-02, -1.5623e-02, -2.0974e-02, -4.8120e-03,
-7.6539e-06, -1.5232e-02, -8.6292e-07, -4.7801e-02, -5.4409e-02,
3.9071e-03, -6.3406e-02, 2.7563e-05, 4.1428e-02, 8.6515e-02,
9.5636e-02, -3.5236e-03, 1.0058e-02, -4.3007e-03, -1.9829e-02,
1.6537e-03, -1.4498e-02, -6.3884e-03, 1.1822e-01, -2.4640e-03,
5.6290e-03, 1.7343e-04, -3.6506e-04, 1.8449e-02, -7.8110e-02,
-1.1054e-01, -1.2492e-02, -2.4699e-06, 2.3808e-03, 1.6511e-03,
-3.6446e-02, -1.9568e-03, -7.9801e-03, -3.1437e-02, -5.7732e-03,
-1.3295e-01, 1.2971e-02, -5.8639e-02, 1.3708e-02, -9.6857e-04,
2.9041e-02, 6.1101e-07, 2.3282e-08, 1.2861e-06, 5.5256e-03,
-1.0595e-02, 1.5939e-02, 2.6585e-03, 2.7292e-07, 3.6939e-03,
1.3843e-01, -3.5164e-02, 3.3416e-02, -1.3455e-02, -1.1674e-01,
1.3319e-08, -4.7397e-02, 1.2603e-02, -4.7765e-02, 3.6308e-02,
-4.5951e-02, 1.3415e-01, -1.0641e-07, -5.2363e-03, -3.6551e-02,
5.5430e-02, -7.1817e-02, 6.0620e-03, 6.9184e-04, 6.7842e-03,
-5.1369e-03, -1.5799e-02, -1.9165e-02, -2.1009e-02, -4.4938e-03,
-5.9774e-03, -4.1721e-08, -3.1079e-03, -6.4795e-03, 1.7523e-02,
7.9426e-02, 2.6991e-02, -1.2950e-01, 6.5436e-03, 1.1590e-02,
-1.1801e-02, 2.8152e-02, 4.0480e-03, -1.1941e-01, 1.7171e-03,
-6.6006e-02, 1.1131e-02, 4.1784e-03, -1.4346e-03, -1.2590e-05,
-2.5132e-02, 3.1090e-02, -1.2760e-01, -1.0714e-01, -5.5529e-02,
6.1587e-07, 3.9569e-03, -4.8457e-02, -2.7235e-02, -2.4397e-02,
-1.2970e-01, 3.9750e-03, 2.3700e-03, 1.6577e-02, -1.3715e-02,
-2.0680e-02, -2.5951e-03, -5.7122e-03, -6.4470e-02, 1.7800e-02,
4.5163e-02, 9.3989e-06, -1.4030e-02, -1.7638e-02, -1.5139e-02,
1.0327e-02, 3.1424e-02, -1.2606e-02, 6.4144e-03, 2.6532e-02,
1.2491e-02, 1.4365e-02, -2.4786e-02, 1.1143e-02, -2.8458e-04,
2.4863e-07, -4.9802e-03, 1.3974e-02, -2.2172e-02, -2.5483e-02,
1.8387e-01, 1.0178e-02, -1.8325e-02, -7.5406e-02, 1.3495e-07,
-5.4790e-02, 1.2497e-02, 5.3769e-03, -1.0004e-07, 2.3629e-02,
-1.3904e-02, -6.2278e-03, -2.2938e-02, 1.1047e-02, -2.7057e-02,
3.3051e-03, 7.9330e-04, 7.6237e-02, -1.3739e-02, 7.9737e-02,
2.2824e-02, 3.4137e-09, 1.1113e-02, -2.4318e-02, -2.4703e-02,
-3.3414e-06, -1.2208e-01, -1.0338e-01, 4.1802e-06, 8.3050e-02,
-3.3793e-02, -5.4347e-03, -1.5232e-02, 8.0727e-02, 3.0390e-02,
1.9241e-02, 5.7994e-03, -1.4920e-02, 2.6332e-02, -1.3213e-02,
-3.0423e-02, 9.9620e-07, 1.6393e-02, -3.2528e-02, 3.7154e-05,
1.7973e-01, 1.2172e-07, 7.2994e-08, 2.8944e-03, -1.3729e-05,
-6.2808e-02, 2.2089e-03, 4.3174e-02, 2.7678e-02, -2.8582e-02,
-6.6965e-02, 4.5367e-02, -1.1172e-03, -1.0364e-02, 6.7318e-03,
-1.0991e-02, 7.1438e-02, -1.3086e-07, -1.0517e-02, -2.5124e-02,
2.4157e-02, 1.9690e-02, -8.9569e-09, 4.6604e-03, 1.0361e-02,
-3.2636e-02, -6.7676e-03, 2.3461e-02, -6.5343e-02, -1.6444e-02,
9.8199e-03, 1.1577e-02, -4.0889e-04, -4.2558e-02, -1.0052e-02,
2.3330e-02, -8.0791e-06, -2.8848e-02, -9.0594e-04, 1.5898e-02,
2.7582e-02, 1.1086e-01, 1.0211e-02, -1.3122e-01, -7.0897e-03,
-7.2518e-03, -3.9014e-02, 3.8611e-02, 2.8432e-08, 7.6105e-03,
1.1481e-02, -4.2232e-03, 1.6695e-02, -2.8884e-02, 4.4396e-03,
5.5876e-02, 1.9523e-04, 2.5766e-02, 1.7733e-02, 1.4922e-08,
-1.0465e-02, -2.3095e-03], device='cuda:0')), ('layer2.0.bn3.running_var', tensor([ 3.7442e-04, 3.3829e-03, 4.3156e-06, 2.3976e-03, 2.1590e-04,
1.0881e-11, 4.0602e-03, 3.8027e-04, 1.8224e-04, 6.1868e-04,
9.8953e-04, 1.7510e-03, 2.0537e-03, 4.0312e-04, 3.7781e-11,
1.8135e-03, 8.9379e-03, 2.1450e-04, 3.7433e-03, 1.3872e-03,
1.7762e-03, 2.5999e-04, 4.2244e-03, 6.5226e-14, 5.2279e-03,
2.3938e-04, 1.7051e-03, 3.5971e-03, 2.5149e-10, 5.2685e-03,
3.4308e-03, 3.2761e-04, 1.5974e-03, 2.0435e-03, 1.7339e-03,
4.7105e-03, 5.5768e-03, 4.9848e-03, 7.9672e-06, 1.5914e-02,
5.9588e-03, 4.5156e-03, 9.7594e-04, 1.9554e-04, 1.6856e-04,
3.9151e-03, 2.5751e-04, 2.7105e-03, 3.3405e-03, 2.3678e-03,
2.9525e-04, 6.5310e-08, 2.6987e-03, 2.0065e-03, 2.7232e-04,
3.5316e-03, 6.7215e-04, 9.6770e-04, 5.9261e-03, 1.3912e-12,
1.5644e-04, 2.7614e-03, 2.1444e-03, 1.8534e-04, 1.7851e-03,
2.4335e-03, 5.6722e-04, 3.5066e-03, 1.4922e-11, 3.6501e-03,
5.2506e-03, 3.0117e-03, 3.9815e-03, 2.3610e-03, 7.0684e-03,
3.7378e-04, 3.3198e-03, 3.2547e-04, 5.5685e-03, 2.1839e-03,
4.5281e-03, 2.0990e-03, 7.4667e-05, 4.1867e-03, 2.0475e-03,
5.7212e-03, 5.0468e-03, 2.0831e-04, 4.5367e-04, 1.9174e-11,
6.3673e-12, 3.3361e-03, 1.3640e-03, 2.9085e-04, 2.0138e-15,
2.2443e-03, 4.7834e-03, 6.0526e-04, 2.1229e-03, 5.3558e-04,
2.4687e-03, 2.4473e-03, 3.8821e-04, 4.4522e-03, 3.7972e-04,
2.0754e-07, 3.8851e-03, 2.0933e-03, 4.1729e-03, 4.5962e-04,
3.2173e-03, 4.9421e-03, 3.3031e-13, 2.9723e-03, 3.1239e-04,
5.2151e-03, 2.3834e-03, 3.1846e-03, 4.2795e-03, 1.6933e-03,
2.0683e-03, 2.3601e-03, 2.1098e-03, 2.3151e-03, 1.1880e-03,
6.2873e-04, 5.4233e-03, 1.9403e-03, 8.6777e-04, 3.0571e-03,
4.2840e-03, 2.5471e-03, 2.3402e-03, 1.9450e-11, 3.6830e-03,
1.1009e-03, 3.5929e-03, 4.0083e-03, 2.4780e-13, 2.7612e-03,
3.1313e-03, 2.8353e-03, 3.6458e-03, 5.5202e-08, 4.1756e-04,
2.5792e-04, 2.2652e-04, 2.6066e-03, 7.0653e-03, 3.3390e-03,
7.5587e-03, 1.4079e-03, 2.1390e-03, 2.9125e-10, 3.2973e-04,
4.1057e-03, 3.0859e-04, 1.6013e-11, 3.9742e-03, 1.4487e-03,
4.9619e-03, 5.0744e-04, 5.0099e-03, 2.3471e-03, 3.4091e-03,
2.3793e-07, 2.6686e-03, 1.6603e-03, 2.8072e-03, 1.3118e-03,
9.5197e-14, 1.3472e-02, 1.5335e-02, 5.3237e-04, 5.4192e-11,
4.7589e-03, 2.1212e-03, 6.8935e-03, 2.3632e-04, 2.1929e-03,
3.0566e-03, 3.5684e-03, 3.8510e-03, 3.9769e-04, 2.3645e-03,
3.0308e-04, 1.0628e-12, 2.3756e-03, 2.8878e-04, 2.4955e-03,
7.1305e-03, 2.6975e-03, 2.3581e-03, 3.0794e-03, 8.4595e-04,
3.7357e-03, 1.5944e-15, 3.9721e-03, 5.1838e-04, 2.9045e-04,
4.1044e-12, 4.6742e-03, 3.9357e-03, 1.3234e-04, 1.4757e-13,
1.2890e-03, 1.7252e-03, 3.0559e-03, 8.7095e-04, 7.9314e-05,
2.1629e-03, 9.9931e-03, 6.4326e-03, 1.9915e-11, 8.3758e-04,
4.0501e-03, 1.5529e-03, 7.6809e-04, 5.2771e-04, 4.9759e-03,
1.0535e-03, 1.5878e-03, 9.3424e-16, 1.3055e-03, 3.9984e-03,
5.0825e-03, 2.6202e-03, 5.8730e-04, 4.1533e-03, 7.6235e-03,
2.3831e-03, 1.8842e-03, 4.2080e-04, 4.4858e-04, 1.9613e-15,
5.6080e-03, 3.1078e-10, 5.1700e-03, 9.0772e-03, 2.3943e-03,
1.5570e-09, 9.7275e-04, 3.1584e-03, 5.1919e-04, 1.5996e-03,
2.3407e-03, 1.6214e-03, 7.2425e-03, 2.3009e-03, 2.7439e-04,
4.8124e-03, 5.8976e-04, 3.3446e-03, 3.5829e-03, 2.1740e-03,
2.8626e-04, 3.3662e-03, 3.5088e-03, 6.6072e-03, 5.5220e-03,
1.3610e-03, 5.6916e-03, 5.7593e-03, 3.5719e-04, 4.2291e-03,
1.2451e-03, 1.5112e-03, 2.0217e-04, 5.3514e-04, 1.8236e-04,
1.0702e-09, 2.4825e-03, 4.2148e-13, 2.8921e-03, 4.6693e-03,
3.5447e-04, 9.9274e-04, 1.9430e-04, 2.4354e-03, 5.0226e-03,
3.1432e-03, 2.5055e-04, 3.2935e-03, 2.0739e-04, 3.6736e-04,
3.1292e-04, 5.8636e-03, 3.3682e-03, 3.9302e-03, 1.4861e-03,
2.3122e-04, 6.9936e-08, 2.9544e-03, 4.2981e-03, 1.5728e-03,
5.7090e-03, 5.8573e-03, 1.5121e-09, 2.2693e-04, 7.5339e-07,
2.8455e-03, 1.8040e-03, 1.3005e-03, 2.2132e-03, 1.4988e-03,
2.2559e-03, 5.9433e-04, 1.8606e-03, 1.0964e-03, 1.5289e-04,
2.6468e-03, 8.4398e-12, 2.4664e-14, 9.9415e-13, 3.0478e-04,
1.4340e-03, 4.0567e-03, 3.9009e-04, 5.4294e-14, 3.0020e-04,
3.1352e-03, 3.8790e-03, 2.1520e-03, 9.0751e-04, 6.7459e-03,
1.9314e-15, 1.9290e-03, 2.8606e-03, 2.5388e-03, 2.4409e-03,
5.3313e-03, 6.7253e-03, 7.7896e-14, 1.4603e-04, 1.4603e-03,
1.5850e-03, 3.5220e-03, 3.3791e-04, 4.2543e-03, 1.1795e-03,
4.0052e-04, 3.4776e-03, 4.2089e-04, 3.7485e-03, 5.7720e-04,
5.9178e-03, 6.6630e-15, 4.8335e-04, 2.3426e-03, 7.9451e-04,
2.8334e-03, 2.2458e-04, 7.9970e-03, 2.3104e-04, 5.4521e-04,
3.8490e-03, 2.2889e-03, 5.3931e-04, 2.6905e-03, 3.8160e-03,
2.8427e-03, 2.8452e-03, 2.0665e-03, 2.6320e-04, 6.4870e-10,
8.2508e-03, 2.6411e-03, 3.7111e-03, 4.2854e-03, 1.9650e-03,
2.9393e-12, 3.7285e-04, 5.6122e-03, 5.1351e-03, 8.0902e-03,
5.9585e-03, 1.5931e-03, 4.5507e-03, 1.3927e-03, 1.6309e-04,
6.3079e-03, 3.6369e-03, 2.3821e-04, 2.2612e-03, 5.0486e-03,
3.4447e-03, 5.3342e-10, 2.9982e-04, 3.5482e-03, 4.7784e-03,
5.4358e-03, 5.3217e-03, 3.8086e-04, 3.2319e-04, 6.3064e-03,
8.2746e-04, 4.3012e-04, 1.7767e-03, 3.4262e-03, 1.3092e-03,
1.1862e-13, 3.8099e-04, 6.0656e-03, 2.2476e-03, 3.5142e-03,
6.7034e-03, 9.4827e-04, 1.7543e-03, 1.9135e-03, 1.2435e-13,
3.1210e-03, 4.1017e-03, 4.0964e-04, 8.4811e-15, 2.0406e-03,
2.4509e-03, 2.5157e-04, 5.6635e-03, 3.0313e-04, 3.8256e-03,
3.3486e-03, 5.1491e-04, 1.4597e-03, 2.8244e-03, 1.4560e-03,
1.1697e-03, 9.1179e-14, 8.8341e-04, 4.1782e-03, 1.3298e-03,
3.9989e-12, 1.4510e-03, 4.9719e-03, 2.3324e-11, 4.4342e-03,
2.4836e-03, 3.0625e-03, 4.1252e-03, 2.8669e-03, 4.7490e-03,
1.4760e-03, 4.1553e-04, 1.4741e-03, 7.7835e-04, 4.6468e-04,
3.3930e-03, 1.4298e-12, 5.6873e-03, 7.1419e-03, 7.7385e-09,
4.1991e-03, 3.4553e-13, 4.0473e-14, 4.7843e-03, 2.8699e-06,
2.5819e-03, 3.8838e-03, 2.7354e-03, 5.8979e-04, 6.1263e-03,
1.3903e-03, 3.5978e-03, 3.1759e-04, 2.9094e-03, 7.1533e-04,
4.8588e-03, 7.6915e-04, 2.3253e-14, 4.0203e-04, 3.5872e-03,
1.8740e-03, 6.3220e-04, 8.1185e-16, 2.7269e-04, 3.4606e-04,
3.0068e-03, 1.2638e-03, 4.6048e-04, 4.9300e-03, 3.9457e-03,
7.5406e-04, 3.6764e-03, 3.1731e-04, 3.7191e-03, 5.6723e-03,
1.8850e-03, 5.6276e-11, 3.1795e-03, 2.8680e-04, 2.0351e-03,
3.8749e-03, 2.0958e-03, 3.5452e-04, 5.4052e-03, 3.8645e-04,
2.9737e-03, 4.9700e-03, 1.0714e-03, 6.1553e-16, 3.1714e-03,
1.7863e-04, 1.3058e-03, 2.8641e-03, 1.2372e-03, 2.6002e-03,
5.2840e-04, 2.3130e-04, 9.5191e-04, 2.1299e-03, 1.2094e-13,
5.6252e-03, 3.7927e-04], device='cuda:0')), ('layer2.0.downsample.0.weight', tensor([[[[ 6.9636e-02]],
[[ 3.6008e-03]],
[[ 1.3478e-02]],
...,
[[ 1.8134e-01]],
[[ 1.9458e-03]],
[[ 3.1590e-03]]],
[[[-1.0936e-02]],
[[-3.4424e-03]],
[[ 1.6514e-02]],
...,
[[ 3.2527e-03]],
[[-3.9140e-02]],
[[-5.9866e-03]]],
[[[ 8.8925e-04]],
[[ 1.0271e-03]],
[[ 1.5778e-03]],
...,
[[ 2.5194e-04]],
[[-2.1439e-03]],
[[-1.1421e-04]]],
...,
[[[ 1.0406e-07]],
[[ 5.3282e-08]],
[[ 9.3874e-08]],
...,
[[-3.8367e-08]],
[[-1.5700e-07]],
[[ 1.0596e-07]]],
[[[ 8.8278e-03]],
[[-2.4494e-03]],
[[-4.0442e-02]],
...,
[[ 1.2038e-02]],
[[ 2.7628e-02]],
[[ 1.4599e-02]]],
[[[ 7.5303e-03]],
[[-4.6605e-02]],
[[-5.5020e-02]],
...,
[[ 1.6583e-02]],
[[-4.7406e-02]],
[[ 1.2077e-02]]]], device='cuda:0')), ('layer2.0.downsample.1.weight', tensor([ 2.5894e-01, 1.0952e-01, 2.4350e-03, 9.8841e-02, 1.4828e-01,
2.2900e-06, 1.2362e-01, 2.5867e-01, 1.7806e-01, 2.6309e-02,
8.7784e-02, 2.2239e-01, 1.9097e-01, 2.2899e-01, -2.7017e-06,
1.1980e-01, 7.9462e-02, 1.2918e-01, 8.1540e-02, 1.6561e-01,
8.5212e-02, 2.0080e-01, 1.7026e-01, 3.6475e-07, 2.3972e-01,
9.7817e-02, 1.0577e-01, 9.8964e-02, -4.4109e-06, 5.2819e-02,
1.9113e-01, 1.5563e-01, 1.0492e-01, 1.6792e-01, 2.1628e-01,
1.4154e-01, 2.2748e-01, 1.5448e-01, -2.0087e-03, 1.4546e-01,
9.9183e-02, 3.7556e-02, 9.3432e-02, 2.4658e-01, 6.6383e-02,
8.1327e-02, 2.4016e-01, 1.4118e-01, 3.5700e-02, 1.1152e-01,
1.4306e-01, 1.0296e-03, 2.9085e-02, 2.2361e-01, 3.1879e-01,
4.0880e-02, 1.4030e-01, 7.4188e-02, 1.3349e-01, 8.4435e-07,
1.4750e-01, 1.4078e-01, 5.6040e-02, 2.3601e-01, 2.6426e-02,
1.0163e-01, 2.7476e-01, 7.1125e-02, 2.8428e-06, 6.3914e-02,
9.2293e-03, 1.2069e-01, 9.6343e-02, 1.0677e-01, 3.5325e-02,
2.7641e-01, 5.7064e-02, 2.2450e-01, 1.6698e-01, 8.4091e-02,
7.4761e-02, 1.1079e-01, 4.9667e-02, 7.8219e-02, 1.3103e-01,
1.2695e-01, 1.2179e-01, 7.4704e-02, 2.9422e-01, -1.7493e-06,
1.9132e-06, 9.9225e-02, 6.5375e-02, 2.0994e-01, 5.8966e-09,
3.3057e-02, 1.6077e-01, 2.2146e-01, 2.2166e-01, 3.6653e-01,
2.8574e-01, 5.7153e-02, 2.7196e-01, 8.0166e-02, 2.1129e-01,
1.2923e-04, 7.7990e-02, 1.0840e-01, 1.6766e-01, 1.7519e-01,
1.9105e-01, 8.9661e-02, 7.1795e-07, 9.4744e-02, 1.9144e-01,
1.5211e-01, 8.0793e-02, 1.7757e-01, 1.0797e-01, 1.7612e-01,
9.4192e-02, 1.4459e-01, 1.9132e-01, 1.9754e-01, 1.1739e-01,
2.7245e-01, 1.0505e-01, 2.1353e-01, 1.1512e-01, 1.0201e-01,
1.0968e-01, 6.4043e-02, 1.3702e-01, 1.5151e-06, 2.1890e-01,
9.4661e-02, 1.2666e-01, 1.8719e-01, -1.3624e-07, 1.0235e-01,
1.7126e-01, 7.2575e-02, 9.3477e-02, 8.6764e-05, 9.3516e-02,
3.4630e-01, 2.9881e-01, 1.3225e-01, 4.2841e-02, 2.5396e-01,
1.2691e-01, 5.6624e-02, 2.1234e-01, 9.4354e-06, 2.4363e-01,
2.1899e-01, 2.4067e-01, 2.0311e-06, 2.0009e-01, 1.1029e-01,
7.6456e-02, 2.2377e-01, 7.9825e-02, 2.8083e-01, 6.8383e-02,
3.1770e-04, 2.3516e-01, 1.0715e-01, 5.0745e-02, 6.1870e-02,
5.0323e-07, 1.8813e-01, 2.2219e-02, 1.8896e-01, 9.1058e-06,
-9.3720e-03, 1.4908e-01, 1.2078e-01, 1.4497e-01, 2.3185e-01,
1.1183e-01, 1.3252e-01, 7.5160e-02, 2.4048e-01, 2.5597e-01,
3.1719e-01, 9.3287e-07, 1.7016e-01, 2.7553e-01, 1.1209e-01,
1.9916e-01, 1.3921e-01, 2.2639e-01, 2.5571e-01, 2.9736e-01,
1.0919e-01, -2.1788e-08, 4.1135e-02, 1.0497e-01, 2.7017e-01,
-1.1039e-06, 4.1734e-02, 1.9707e-01, 9.7280e-02, 2.5079e-07,
8.4855e-02, 5.4263e-02, 1.5409e-01, 2.3644e-01, -3.3276e-03,
2.0224e-01, 8.1634e-02, 1.8215e-01, 2.9152e-06, 6.7323e-02,
8.5177e-02, 1.6566e-01, 2.6869e-01, 4.0581e-02, 6.9725e-02,
1.6717e-01, 6.8251e-02, 4.3876e-08, 6.3462e-02, 1.8687e-01,
5.9377e-02, 9.9933e-02, 2.7918e-01, 8.4557e-02, 6.4699e-02,
3.1337e-02, 2.1270e-01, 1.6158e-02, 4.6383e-02, 4.3892e-08,
5.5247e-02, 7.5510e-06, 6.5603e-02, 5.0101e-02, 2.0750e-01,
6.1390e-05, 3.1456e-01, 4.8928e-02, 2.3551e-01, 4.5786e-02,
1.4947e-01, 2.8260e-01, 1.0210e-01, 7.0082e-02, 1.9167e-01,
4.2421e-02, 2.1541e-01, 7.3094e-02, 1.2241e-01, 4.8193e-02,
2.0008e-01, 5.0354e-02, 1.3672e-01, 3.6754e-02, 1.3625e-01,
1.2990e-01, 5.0630e-02, 5.2546e-02, 2.3381e-01, 6.6546e-02,
2.4019e-01, 3.0668e-02, 3.0689e-01, 2.2720e-01, 2.3522e-01,
4.6171e-05, 8.1741e-02, 8.7562e-07, 1.3516e-01, 6.1836e-02,
2.0586e-01, 1.3580e-01, 8.8719e-02, 6.4561e-02, 1.2725e-01,
1.7426e-01, 1.8437e-01, 7.7724e-02, 1.5262e-01, 1.7132e-01,
3.1335e-01, 3.4451e-02, 1.2605e-01, 1.2839e-01, 1.0430e-01,
2.7461e-01, 3.5598e-04, 5.6617e-02, 8.7690e-02, 1.5311e-01,
1.1519e-01, 4.3404e-02, 4.2354e-05, 2.3621e-01, -7.8637e-06,
8.4913e-02, 1.4516e-01, 5.9520e-02, 1.1899e-01, 1.5344e-01,
1.2991e-01, 6.9376e-02, 1.4720e-01, 1.5015e-01, 3.4827e-01,
7.8291e-02, 2.7767e-06, 6.0657e-08, -2.3011e-07, 2.5938e-01,
8.8719e-02, 3.8580e-02, 3.3758e-01, 2.0945e-07, 1.4973e-01,
6.0445e-02, 2.0537e-01, 8.5454e-02, 7.9034e-02, 1.6697e-01,
-2.9826e-08, 1.3037e-01, 1.7994e-01, 1.7999e-01, 9.8017e-02,
1.0569e-01, 6.9529e-02, 1.5887e-07, 3.4035e-01, 8.6399e-02,
1.6570e-01, 1.1431e-01, 2.3633e-01, 6.3789e-02, 1.7756e-01,
2.2179e-01, 9.1697e-02, 2.5909e-01, 2.8606e-02, 2.4261e-01,
1.7812e-01, 1.1581e-08, 1.8174e-01, 6.3036e-02, 2.2299e-01,
2.1315e-01, 2.1711e-01, 1.6508e-01, 2.9226e-01, 2.0710e-01,
1.5458e-01, 1.4579e-01, 2.2657e-01, 2.0913e-01, 5.4199e-02,
1.4851e-01, 2.0314e-01, 1.0949e-01, 3.2031e-01, 1.3787e-05,
1.8286e-01, 8.8698e-02, 9.3633e-02, 2.0066e-01, 1.6375e-01,
1.7920e-06, 2.1021e-01, 1.5132e-01, 1.2822e-01, 1.8261e-02,
9.5773e-02, 1.3455e-01, 9.0493e-02, 5.6843e-02, 2.1273e-01,
-1.9481e-02, 1.1177e-01, 2.0018e-01, 1.4718e-01, 8.9305e-02,
6.1995e-02, 2.2648e-05, 2.6067e-01, 2.3732e-01, 9.6725e-02,
2.7169e-02, 2.7572e-02, 1.5351e-01, 1.1330e-01, 1.2615e-01,
3.1682e-01, 3.6433e-01, 1.5977e-01, 1.3429e-01, 5.2137e-02,
4.6438e-07, 3.5380e-01, 2.2400e-02, 2.0857e-01, 2.3714e-01,
2.0609e-01, 2.7071e-01, 1.1499e-01, 1.0280e-01, 3.3806e-07,
1.2464e-01, 7.5613e-02, 1.7810e-01, 5.9839e-08, 9.7862e-02,
4.7789e-02, 9.8830e-02, 1.3672e-02, 2.6721e-01, 1.3875e-01,
9.9412e-02, 5.8001e-02, 8.8942e-02, 1.8603e-01, 2.2661e-01,
5.3778e-02, 2.9870e-07, 7.1578e-02, 1.6625e-01, 1.3182e-01,
2.2102e-06, 6.4613e-02, 4.4592e-02, 1.4840e-06, 3.7320e-01,
8.3566e-02, 3.6263e-02, 1.2490e-01, 1.4024e-01, 6.7643e-02,
8.7781e-02, 2.4787e-01, 2.7908e-01, 2.9175e-01, 1.0252e-01,
8.8438e-02, -2.4200e-07, 8.3925e-02, 1.8358e-01, -2.7752e-05,
6.4146e-02, 5.6049e-07, 1.0697e-07, 5.9478e-02, 9.6309e-04,
2.5316e-01, 5.5721e-02, 1.0059e-01, 3.5312e-01, 2.6611e-01,
4.9717e-02, 3.2800e-02, 2.8730e-02, 1.9604e-01, 2.3787e-01,
9.8731e-02, 2.1813e-02, 7.8078e-08, 2.5368e-01, 6.4311e-02,
1.0932e-01, 3.2254e-01, 1.4187e-08, 2.1543e-01, 1.8553e-01,
1.2584e-01, 2.2427e-01, 1.3510e-01, 1.3744e-01, 2.8398e-02,
1.1102e-01, 1.4465e-01, 9.9769e-02, 1.4012e-01, 7.7053e-02,
1.0650e-01, -1.2151e-06, 1.5623e-01, 1.8287e-01, 2.6775e-01,
4.6327e-02, 8.6280e-02, 1.4811e-01, 8.9865e-02, 2.9692e-01,
2.4645e-01, 9.5585e-02, 1.9333e-01, 1.5000e-08, 1.2984e-01,
1.4060e-01, 2.0470e-01, 5.6103e-02, 1.4413e-01, 8.3059e-02,
2.8040e-01, 2.5632e-01, 1.4756e-01, 1.4561e-01, 1.8631e-07,
7.1401e-02, 2.6416e-01], device='cuda:0')), ('layer2.0.downsample.1.bias', tensor([ 5.0408e-02, -5.3801e-02, -4.9714e-03, 2.0539e-02, 4.7344e-02,
-7.9189e-06, 9.5915e-03, 8.5763e-02, 3.1474e-02, 7.3589e-02,
3.1398e-03, -3.0815e-02, 4.4212e-03, 9.0432e-02, -1.7023e-05,
-2.1265e-02, -3.0664e-02, 3.1546e-02, 4.3062e-02, -7.6598e-04,
4.0609e-02, 7.0054e-02, -3.2688e-03, -8.6219e-07, 2.9686e-02,
6.5213e-02, 3.7850e-03, -3.1191e-02, -3.1902e-05, 4.1727e-02,
-9.7415e-02, 7.2782e-02, 3.0331e-02, -1.1470e-02, 8.0664e-03,
-2.3092e-02, 9.3858e-02, 4.7561e-02, -4.6917e-03, 7.5247e-02,
-2.9921e-02, 1.5897e-02, -2.7721e-02, 4.7655e-02, 7.5995e-03,
-5.5337e-02, 7.0081e-02, -3.3844e-02, 6.6372e-02, -5.6138e-02,
9.5078e-02, -8.3608e-04, 5.9420e-02, 4.6229e-02, 9.6871e-02,
1.0570e-02, -1.0752e-02, -3.7556e-03, 1.1314e-02, -4.0713e-06,
4.4267e-02, -3.9143e-03, -7.9898e-03, 3.3193e-03, -1.0885e-02,
3.3494e-03, -9.2455e-02, 3.9653e-02, -7.8888e-06, -2.4530e-03,
1.3655e-02, 1.8437e-02, -1.1777e-02, -1.0059e-02, -4.4714e-02,
5.1907e-02, -2.4790e-02, 1.0142e-01, 1.8356e-02, -5.6835e-02,
-1.9470e-04, 1.7989e-02, -2.3642e-02, 2.5737e-02, -2.2398e-03,
4.8911e-02, 5.5182e-02, -1.8045e-03, 8.5885e-02, -1.0144e-05,
-7.6508e-06, -4.7996e-02, -5.4346e-02, 3.8387e-02, -1.5890e-07,
-1.3694e-02, 2.1481e-02, 3.7525e-02, -3.2900e-03, -1.4641e-02,
-5.8719e-03, -3.0566e-02, 9.1605e-02, 2.9773e-02, 1.0682e-01,
-8.9958e-04, 3.5209e-02, -4.2207e-02, -3.2497e-03, -3.2672e-02,
8.3064e-02, 4.8578e-02, -2.8526e-06, -1.8117e-02, 1.0543e-01,
7.2242e-02, -7.6185e-03, 2.2616e-03, 3.7346e-02, 4.6880e-02,
-7.9570e-03, 3.6065e-02, 7.7348e-02, 3.2475e-03, -3.0122e-02,
1.5840e-01, -3.3549e-02, 5.9058e-02, -5.4857e-04, -4.5232e-02,
-2.0058e-02, 1.7915e-02, 1.7743e-02, -7.1322e-06, 1.1004e-01,
-7.1586e-03, -7.0238e-04, 3.5450e-03, -1.4194e-06, -5.0723e-03,
1.0719e-02, -2.6846e-02, -1.5724e-02, -4.8029e-04, -4.1229e-02,
1.0429e-02, 4.7818e-02, -2.1555e-02, 2.1728e-02, -2.0473e-02,
2.5582e-05, 3.2087e-02, 6.4179e-02, -6.2677e-05, 1.1763e-01,
3.7077e-02, 8.7900e-02, -9.1733e-06, 3.5007e-02, -1.5780e-02,
-1.4645e-02, 1.1493e-01, 3.5553e-02, 3.8060e-02, 5.2214e-02,
-1.3884e-03, -1.9273e-02, -3.3458e-02, -3.7974e-02, -1.7768e-02,
-1.3500e-06, 7.1225e-03, 9.3039e-02, 1.4888e-01, -2.7238e-05,
-5.6481e-02, 1.0640e-01, -3.8111e-02, 5.7626e-02, -1.1871e-02,
-3.7452e-02, -8.3534e-03, 1.3115e-02, 5.2240e-02, 1.4609e-02,
1.3691e-01, -2.8239e-06, 5.4831e-03, 7.0706e-02, -3.6691e-02,
-1.7748e-03, -1.1044e-02, 2.9012e-02, 1.1499e-01, 1.5405e-01,
7.9285e-03, -1.1732e-07, -2.6413e-02, 1.2214e-02, 8.3662e-03,
-6.2902e-06, 5.0012e-02, -6.5154e-03, 5.0267e-02, -7.1198e-07,
-4.5093e-02, 2.5603e-02, 3.9757e-02, 4.9278e-02, -9.7368e-03,
-1.0309e-03, 8.0354e-02, 4.9503e-02, -1.0135e-05, -2.0469e-02,
-2.3065e-02, -1.0967e-02, 1.3992e-01, 1.3652e-02, 2.2791e-02,
5.5029e-02, -1.3392e-02, -1.2968e-07, -3.2554e-02, 2.6425e-02,
2.0118e-02, 5.8526e-02, 1.9821e-01, 2.1531e-03, 1.3833e-02,
-1.2165e-02, 7.6226e-02, 1.0087e-01, -2.9128e-02, -1.1165e-07,
3.4191e-02, -2.7389e-05, -6.7306e-03, 5.2081e-02, 5.8233e-03,
-1.8436e-04, 1.9059e-01, 2.4796e-02, 8.9079e-02, -2.8740e-03,
-5.1246e-02, 2.3442e-02, 1.3485e-04, -4.4533e-02, 3.7772e-02,
4.5265e-02, 1.2982e-01, -7.0863e-03, 2.9466e-03, 2.1229e-02,
6.8678e-02, -2.4108e-02, -5.9673e-03, 3.3583e-02, 5.1814e-03,
-3.4058e-02, 2.4528e-02, 3.2980e-03, 8.5338e-02, 4.6658e-03,
1.1495e-01, 2.1275e-02, 2.0298e-02, 6.5352e-02, 4.8504e-02,
-1.2237e-04, -2.2201e-02, -3.2690e-06, 5.6547e-02, 9.2143e-03,
5.3177e-02, 1.3121e-02, 5.8778e-02, -1.0628e-02, -4.6647e-02,
-3.8120e-03, 1.6281e-02, -1.6294e-02, 4.6992e-02, 8.2434e-02,
-3.7641e-02, 9.0587e-03, -2.1060e-03, -2.2506e-02, -4.0034e-02,
-1.7160e-02, -9.8877e-04, -4.3668e-02, -4.3654e-02, -7.9761e-02,
8.0995e-04, 3.3363e-02, -9.5408e-05, 2.4635e-02, -1.2186e-03,
8.8857e-03, -2.5290e-02, 1.5415e-02, 4.1727e-02, -1.4499e-02,
-2.5007e-03, -4.2352e-02, 2.3150e-02, 1.0252e-02, -5.0827e-02,
-2.9843e-02, -1.7975e-05, -2.4256e-07, -2.3200e-06, 8.7960e-02,
-1.9594e-02, -1.4285e-02, 1.3767e-01, -5.4095e-07, 7.2719e-02,
1.5425e-02, 2.2584e-03, 1.6702e-02, -1.8719e-03, 4.0983e-02,
-1.5115e-07, -6.9896e-02, -1.1585e-02, -3.8220e-03, 3.9209e-02,
7.9392e-02, -3.1275e-03, -6.0793e-07, -6.5054e-02, -4.6122e-03,
2.2881e-02, -1.5298e-02, 7.1332e-02, -2.5570e-02, -2.5730e-02,
1.0177e-01, -3.8346e-03, 1.1747e-01, -3.2341e-02, 1.2023e-01,
9.6849e-03, -1.8967e-07, 9.6480e-02, -3.4635e-03, 1.3868e-01,
-7.0695e-04, -9.4783e-02, -4.2773e-03, 3.6939e-02, 1.4341e-01,
8.6191e-03, 3.4905e-02, 1.3142e-01, 2.1686e-02, -3.2155e-02,
2.2250e-02, 3.3670e-02, 1.1841e-03, -4.6307e-02, -6.7543e-05,
-1.9290e-02, -1.0294e-03, -2.2324e-02, 6.6470e-02, -2.0465e-02,
-5.0082e-06, 1.1699e-01, 3.5496e-02, 2.2371e-02, 7.6098e-02,
-3.1274e-02, -2.6378e-02, 4.4431e-03, -1.3276e-02, 5.5820e-02,
-2.9341e-02, -3.4241e-04, 1.2948e-01, 4.8526e-02, 1.2148e-03,
1.5532e-02, -5.9827e-05, 9.0663e-02, 6.6560e-02, -2.1857e-02,
-4.2524e-04, 8.4397e-04, 7.5189e-02, -4.0382e-02, -5.7350e-03,
1.3059e-01, 8.9209e-02, -5.2390e-02, 8.3347e-03, 3.6906e-02,
-1.9198e-06, -2.5754e-02, -1.8060e-02, 1.5707e-02, -7.2462e-02,
-5.3932e-02, 3.3604e-02, 5.3753e-03, 9.2454e-03, -9.6343e-07,
1.1055e-02, 9.3140e-03, 1.2375e-01, -2.4052e-07, 4.7903e-03,
-2.4964e-02, 7.8106e-03, -2.5364e-02, -1.7556e-02, 1.1134e-02,
-1.5845e-02, -4.8787e-02, 1.7304e-03, -1.9681e-02, -1.2093e-03,
-1.8620e-02, -1.9253e-06, 6.8844e-03, -1.2531e-02, -2.6413e-04,
-8.7646e-06, 4.8010e-02, 8.7525e-02, -8.0243e-06, -1.7114e-01,
-1.3433e-02, -4.0059e-02, -5.1796e-02, 6.7273e-03, 7.6271e-03,
1.7412e-02, 7.1451e-02, -6.3535e-02, 1.4390e-01, 5.2225e-02,
-2.6857e-02, -2.3525e-06, 9.4138e-03, -2.4442e-03, -1.6837e-04,
-2.2859e-02, -2.3380e-06, -4.7066e-07, 2.9032e-02, -4.4317e-03,
1.2823e-02, 6.2972e-02, -2.8694e-02, 1.0436e-01, 1.2098e-01,
-6.8384e-03, 6.1803e-02, -2.1682e-02, 3.3520e-02, 1.8442e-01,
2.0452e-02, -2.2529e-02, -3.8024e-07, 4.7503e-02, -1.6849e-02,
-3.5622e-02, 1.3090e-01, -1.0422e-07, 6.6521e-02, 8.8123e-02,
1.2963e-02, 6.9495e-02, 2.2898e-03, -8.1541e-03, -5.9840e-02,
-3.7383e-03, 1.6682e-02, 1.8710e-02, -3.3964e-02, -6.9819e-03,
-8.8785e-03, -1.6550e-05, 3.6695e-02, 7.4770e-02, -3.6443e-03,
-2.7702e-02, -5.7924e-02, 5.3831e-02, 3.8519e-02, 9.7022e-02,
-4.2779e-02, -1.2827e-02, -5.4793e-02, -6.0001e-08, 7.1382e-03,
6.7838e-02, -9.0207e-02, -2.3035e-02, -8.5935e-04, 2.0521e-02,
7.8742e-02, 3.6046e-02, -7.2988e-03, -4.1833e-02, -8.7138e-07,
7.5300e-02, 1.6350e-01], device='cuda:0')), ('layer2.0.downsample.1.running_mean', tensor([-3.0674e-02, -1.8816e-02, -6.0806e-03, 5.2650e-02, 6.3695e-02,
-8.8839e-06, -8.3040e-02, 9.8298e-03, 1.5734e-01, -2.3900e-02,
-3.7858e-02, -1.1380e-01, -8.3222e-02, -1.0196e-01, -5.6258e-06,
-1.0480e-01, 1.2350e-01, 8.0340e-02, -3.5935e-02, 9.5215e-03,
5.1797e-04, -1.7991e-02, -8.5454e-02, -6.3344e-07, 9.1694e-03,
-7.0159e-02, -5.5278e-02, -1.9926e-02, 1.4802e-05, -6.2171e-02,
3.9130e-02, -2.2871e-01, 1.0833e-02, -1.7398e-01, 8.3256e-02,
-4.9231e-03, -1.6027e-01, -6.0657e-03, -6.0056e-03, -6.5826e-02,
-2.5781e-02, -6.9274e-03, -1.1741e-01, -1.0807e-01, 9.4167e-02,
-4.5849e-02, 3.3790e-01, -7.5234e-03, 5.5825e-02, 4.4219e-02,
4.7049e-02, 9.7892e-04, 2.6257e-02, -1.7098e-01, -6.6952e-02,
2.3386e-02, -1.9489e-02, -2.8912e-02, -1.8314e-01, 3.6472e-06,
1.5210e-01, 4.1338e-02, 3.0462e-02, 3.0936e-01, 6.0993e-02,
-1.4712e-02, -2.4643e-02, -9.1271e-02, 1.0123e-05, -8.3027e-03,
-4.7268e-02, -5.2653e-02, -1.4295e-01, -5.3757e-02, -4.0290e-02,
-4.2921e-01, -2.3134e-02, -5.7347e-02, -7.2746e-02, 1.8842e-02,
9.1028e-02, 5.7572e-02, -4.7931e-02, 3.2969e-02, -1.0707e-01,
-9.3183e-02, -7.1938e-02, -3.1549e-02, 1.2295e-01, 4.5674e-06,
-8.8228e-06, 1.5532e-02, -3.3382e-02, 1.3533e-01, 1.3728e-07,
-6.7494e-02, 1.7538e-01, 1.8994e-01, 2.6594e-01, 1.8805e-01,
1.1410e-01, -1.2162e-02, 1.2617e-01, 1.0489e-01, -4.6129e-02,
-4.4149e-04, 4.8005e-02, -9.4787e-02, 5.1292e-02, 1.7504e-03,
-3.4595e-01, 7.2662e-02, -2.3841e-06, -7.1231e-02, -6.5081e-02,
1.2484e-01, 1.2121e-01, -1.9372e-01, -1.2332e-01, -5.6733e-02,
-8.5465e-02, -3.1860e-02, 1.2315e-02, -3.9317e-03, -4.7457e-02,
-1.3494e-01, -1.3933e-01, -9.1513e-02, -7.5424e-02, -6.5693e-02,
-7.5887e-02, 3.7968e-03, 3.3925e-02, 3.2874e-06, -1.7255e-01,
-1.1044e-01, -7.4539e-02, -4.5566e-02, 1.8299e-06, -3.4754e-04,
-5.7691e-02, -4.2107e-02, -3.0837e-02, -5.2076e-05, -1.7520e-02,
-2.4840e-01, -2.2332e-02, -6.7793e-02, 3.0758e-02, -1.0791e-01,
5.2236e-02, -6.3962e-02, -2.5805e-02, 5.8996e-05, 6.0256e-02,
-7.2054e-02, 1.1707e-01, -3.6560e-06, 8.0112e-02, 6.4334e-02,
9.1936e-02, -2.4496e-01, -2.7991e-02, 1.2478e-02, -4.0694e-02,
1.6956e-03, -3.6565e-02, -2.3156e-02, 5.6347e-02, 1.1745e-02,
1.6847e-07, -7.2659e-02, -7.0435e-02, -2.7002e-02, -1.9213e-05,
1.3578e-02, -1.5056e-01, -6.2039e-02, 1.5784e-01, -6.6605e-02,
1.2767e-01, -7.7053e-02, 3.6109e-03, 1.0252e-01, -4.5492e-02,
-1.0153e-01, -4.2322e-06, -6.9134e-03, 1.0814e-02, 1.6341e-01,
-1.3720e-01, 2.5687e-02, 1.8300e-02, -2.0035e-01, 1.0777e-01,
-8.2130e-02, 2.6298e-08, -4.1295e-03, -6.2743e-02, 2.2651e-01,
3.3338e-06, 1.3206e-03, 2.4167e-03, -7.9370e-02, -9.3741e-07,
-1.7204e-02, -3.0699e-02, -1.1429e-01, -1.7696e-01, -1.6796e-02,
6.3470e-02, 1.2633e-02, -2.5588e-01, -9.3911e-06, -1.8739e-02,
4.1966e-02, -1.1515e-01, -1.2705e-01, -1.4809e-02, -5.2117e-02,
6.4905e-03, -3.7037e-02, -2.3470e-07, -1.9648e-02, -1.8091e-02,
-4.3771e-02, 6.1556e-02, -6.8652e-02, 8.4727e-02, -1.0230e-01,
8.0167e-03, 3.1332e-02, -7.7673e-02, -3.7713e-02, -1.2623e-07,
7.4961e-02, -4.1733e-05, -8.7472e-02, -4.2894e-02, -1.6639e-01,
-7.0265e-05, -2.1521e-01, 6.0222e-02, -1.1349e-01, -1.3537e-02,
-7.4447e-02, 1.6911e-01, -3.9631e-02, -6.8845e-02, 4.2225e-02,
9.1753e-03, -2.2240e-01, -3.4423e-02, -8.9126e-02, -2.0688e-02,
1.5626e-01, 5.1282e-02, -3.5246e-02, 6.2343e-02, -1.7059e-01,
-4.9905e-02, 8.2088e-02, -2.2199e-02, -2.5447e-01, 1.4360e-02,
-1.0658e-01, -3.4507e-02, -1.0705e-01, 3.2685e-02, 3.2549e-02,
2.3940e-04, -4.6739e-02, 1.1251e-07, 4.8114e-02, -2.6751e-02,
-1.3204e-01, -1.8442e-02, -2.5461e-02, -8.3320e-02, -2.7352e-02,
-4.3390e-02, -7.2617e-02, 3.1021e-02, 1.2757e-01, -7.8584e-02,
6.1902e-02, -2.8250e-02, 4.3275e-02, -8.0131e-02, -5.3454e-02,
4.5607e-02, -1.5365e-03, -1.3086e-02, 5.5492e-02, -1.1524e-02,
-2.8398e-02, 2.3426e-02, 9.5158e-06, 1.1686e-01, 2.5677e-04,
2.3100e-02, -1.3572e-01, -4.1807e-02, -1.1659e-01, -2.0953e-01,
-1.6046e-01, 2.7249e-02, -8.0860e-02, 5.5149e-02, 1.5605e-01,
-9.7663e-02, -2.9863e-06, -3.2524e-07, 1.2546e-06, -4.0618e-02,
6.3032e-02, -3.1097e-02, -1.8536e-01, -2.2208e-07, 8.6735e-02,
-5.1617e-02, 7.8529e-02, 6.8193e-02, -5.9101e-02, -8.4685e-02,
1.2699e-07, -1.4072e-01, -4.3847e-04, 2.6914e-03, -3.4992e-02,
-1.0113e-01, 7.6270e-02, -2.2689e-06, -1.4442e-02, -1.2899e-01,
-7.4108e-02, 9.5515e-02, 9.5323e-03, 5.6982e-02, 6.1278e-02,
-8.8261e-02, 1.2142e-01, -4.9386e-01, -3.7399e-03, 7.7592e-02,
-9.1079e-03, -5.8476e-08, -3.2275e-01, -3.2468e-02, 1.6321e-01,
-1.2479e-02, -1.5101e-02, -7.6152e-02, -2.2486e-02, -1.3466e-01,
-1.1577e-01, 4.2189e-02, -2.7225e-01, -1.2240e-02, 8.2922e-03,
-1.2907e-01, -1.5706e-01, -9.1045e-02, 6.9793e-02, -1.2276e-04,
6.4642e-03, -7.7619e-02, 3.7307e-02, 3.1556e-02, -1.2598e-01,
2.0013e-06, -3.0838e-01, 3.1038e-02, 1.1097e-01, 5.3786e-02,
2.0077e-01, 7.4915e-02, -6.5734e-02, -1.9911e-02, -6.5657e-02,
3.9151e-02, -8.2709e-02, -2.3970e-01, 1.1787e-01, -4.6082e-02,
1.6958e-02, -4.2067e-05, 7.6546e-02, -1.9417e-01, -1.4430e-01,
2.7077e-02, -7.9150e-02, 9.2644e-02, 7.9268e-02, -8.9184e-02,
2.2642e-02, -7.5476e-02, 6.1128e-02, -5.9543e-02, -2.3589e-02,
8.6397e-07, -4.1069e-02, 3.5380e-02, -1.4847e-01, -6.5692e-02,
-3.8401e-02, -3.2387e-02, -2.7683e-02, -9.3460e-03, 2.1284e-06,
-9.8334e-02, -1.6058e-01, -1.3910e-01, -1.9163e-07, 3.2797e-02,
-7.2353e-03, -4.7997e-02, 2.7484e-02, -5.2252e-02, -7.7488e-02,
2.1333e-02, -8.4435e-02, -2.1918e-02, -1.0104e-01, -6.8902e-02,
-2.0636e-02, -9.9021e-07, -4.2065e-02, 1.2234e-02, -1.0143e-01,
-7.1667e-06, -8.4138e-02, -5.0087e-02, -2.4539e-06, 1.2595e-01,
-6.2807e-03, -1.1864e-02, -2.9606e-02, -3.6000e-02, -5.9111e-02,
-4.3415e-02, -3.7978e-02, 6.9282e-02, -1.0479e-01, 1.9267e-02,
-9.1413e-02, -1.1295e-06, 1.8869e-02, -7.1063e-02, 4.1835e-05,
8.2344e-02, 4.4825e-07, 9.4608e-08, -2.6481e-02, -4.0154e-03,
-2.5801e-04, -4.6003e-02, 1.9737e-02, 2.3147e-01, -3.3169e-01,
3.5382e-03, -4.5119e-03, -4.8568e-02, -5.2100e-02, -7.7748e-02,
-4.7613e-02, 1.2548e-02, -1.9980e-08, -1.6176e-01, 4.1314e-02,
2.2528e-03, -4.0230e-02, 4.7711e-08, 1.6076e-01, -7.4929e-02,
-8.1739e-02, -1.2954e-01, 8.5874e-03, 7.1373e-02, 3.3598e-02,
1.1325e-01, -1.6367e-01, 1.8764e-01, -7.4487e-02, -1.3432e-01,
1.0721e-01, -1.3471e-05, -9.5615e-02, -2.1257e-01, -2.7993e-01,
5.8863e-02, 9.1429e-02, -2.4592e-02, 7.3048e-02, -1.7143e-01,
-3.9901e-02, -5.3554e-02, 7.7790e-02, 1.0208e-10, -1.6111e-01,
-4.3970e-02, -3.7334e-02, 4.2446e-02, -1.1351e-02, 2.1408e-02,
-4.2445e-02, 9.3300e-02, -9.7250e-02, 7.1481e-02, -4.6662e-07,
4.9426e-03, -2.4391e-01], device='cuda:0')), ('layer2.0.downsample.1.running_var', tensor([ 3.3956e-02, 5.2252e-03, 6.2095e-05, 7.8726e-03, 1.6759e-02,
7.8198e-11, 1.2341e-02, 4.8789e-02, 1.5406e-02, 2.4308e-03,
2.6470e-03, 1.0681e-02, 1.6049e-02, 2.0288e-02, 3.5680e-11,
5.9570e-03, 6.5520e-03, 9.7892e-03, 7.1140e-03, 1.0325e-02,
7.0225e-03, 1.9240e-02, 2.0923e-02, 1.3765e-12, 2.5838e-02,
7.1027e-03, 5.0059e-03, 8.3978e-03, 1.7524e-10, 5.7648e-03,
1.0329e-02, 2.1835e-02, 1.2189e-02, 7.6186e-03, 1.3340e-02,
1.1866e-02, 6.3229e-02, 1.2301e-02, 1.9408e-05, 4.2031e-02,
7.6161e-03, 2.6368e-03, 3.3348e-03, 1.9533e-02, 3.4843e-03,
3.6164e-03, 1.4341e-02, 6.4147e-03, 2.6219e-03, 5.7463e-03,
2.2538e-02, 2.1927e-06, 3.6462e-03, 1.7934e-02, 2.3656e-02,
2.1460e-03, 7.2612e-03, 3.2556e-03, 1.2822e-02, 2.6570e-11,
1.4542e-02, 1.1122e-02, 2.7355e-03, 1.7353e-02, 1.6654e-03,
4.8245e-03, 8.0124e-03, 5.0658e-03, 1.0613e-10, 2.3041e-03,
1.0507e-03, 1.2834e-02, 5.8100e-03, 5.2017e-03, 1.9466e-03,
1.2355e-02, 1.9132e-03, 3.4085e-02, 2.1364e-02, 3.1422e-03,
6.4746e-03, 6.5949e-03, 1.5980e-03, 4.7580e-03, 8.3022e-03,
1.3151e-02, 1.0991e-02, 7.4892e-03, 6.1516e-02, 1.8955e-11,
5.6978e-11, 5.6800e-03, 3.0695e-03, 2.1070e-02, 9.4738e-15,
1.8880e-03, 1.5728e-02, 1.7340e-02, 1.2486e-02, 7.1939e-02,
1.5117e-02, 1.3906e-03, 4.1910e-02, 6.8132e-03, 3.7582e-02,
1.2493e-06, 7.3211e-03, 3.6420e-03, 2.0574e-02, 1.0973e-02,
2.3432e-02, 1.2754e-02, 3.7478e-12, 5.9241e-03, 1.9196e-02,
3.3924e-02, 5.4891e-03, 1.3118e-02, 1.0137e-02, 1.9379e-02,
4.8932e-03, 1.3588e-02, 2.3132e-02, 1.9240e-02, 6.0142e-03,
7.2515e-02, 1.0700e-02, 2.6646e-02, 5.4666e-03, 7.1055e-03,
1.0353e-02, 5.1299e-03, 1.1792e-02, 2.1949e-10, 4.6736e-02,
5.6747e-03, 1.0756e-02, 1.2728e-02, 6.1225e-13, 4.4281e-03,
2.0149e-02, 5.3588e-03, 5.4679e-03, 5.6299e-08, 3.8849e-03,
1.3792e-02, 1.6157e-02, 1.0008e-02, 3.5088e-03, 1.6360e-02,
1.1453e-02, 5.6142e-03, 2.3529e-02, 2.4949e-09, 3.1352e-02,
3.5178e-02, 3.4717e-02, 5.4637e-11, 1.8298e-02, 6.1990e-03,
4.4439e-03, 2.9325e-02, 1.0493e-02, 2.7259e-02, 2.9528e-03,
3.7642e-06, 2.0646e-02, 5.9019e-03, 2.1828e-03, 2.7699e-03,
1.8256e-12, 5.0421e-02, 4.0329e-03, 2.9544e-02, 8.2690e-10,
7.7376e-04, 1.6300e-02, 1.3702e-02, 1.6112e-02, 1.2497e-02,
4.7302e-03, 1.3904e-02, 5.0829e-03, 4.5389e-02, 1.6714e-02,
2.9825e-02, 2.2512e-11, 1.8238e-02, 2.4738e-02, 1.0666e-02,
2.6866e-02, 1.0297e-02, 2.0770e-02, 5.4295e-02, 1.0335e-01,
1.0483e-02, 3.9217e-15, 1.6940e-03, 5.5238e-03, 3.8060e-02,
3.8709e-12, 5.3157e-03, 1.3253e-02, 4.3258e-03, 1.6342e-12,
2.0583e-03, 1.9300e-03, 1.4019e-02, 6.5277e-02, 6.2814e-05,
1.2943e-02, 1.6020e-02, 3.3828e-02, 7.9091e-11, 2.6202e-03,
4.7781e-03, 1.2484e-02, 5.7681e-02, 4.5571e-03, 4.4597e-03,
1.6626e-02, 3.9569e-03, 6.9870e-15, 1.7055e-03, 2.4042e-02,
4.4846e-03, 8.0530e-03, 8.3862e-02, 7.9232e-03, 6.3318e-03,
1.8823e-03, 2.3692e-02, 1.7321e-03, 6.6349e-04, 7.8351e-15,
7.0483e-03, 2.3019e-09, 6.8160e-03, 5.1863e-03, 1.5196e-02,
1.7166e-08, 1.1528e-01, 3.1884e-03, 1.9729e-02, 2.0040e-03,
9.9624e-03, 1.7449e-02, 1.0284e-02, 2.9304e-03, 1.7480e-02,
2.1710e-03, 4.3223e-02, 4.4851e-03, 9.3167e-03, 3.1914e-03,
1.5994e-02, 2.5323e-03, 1.3045e-02, 4.0699e-03, 1.6281e-02,
3.7098e-03, 4.6410e-03, 3.8154e-03, 2.4499e-02, 5.3729e-03,
3.0536e-02, 3.1496e-03, 1.9790e-02, 3.3996e-02, 1.7345e-02,
2.4818e-08, 2.9927e-03, 9.0562e-12, 2.2459e-02, 4.4485e-03,
3.6513e-02, 1.0910e-02, 5.4677e-03, 4.3805e-03, 1.1111e-02,
1.0444e-02, 1.5234e-02, 4.1500e-03, 1.6526e-02, 2.4650e-02,
3.2881e-02, 2.2071e-03, 7.7526e-03, 6.5113e-03, 3.3519e-03,
2.7395e-02, 6.8161e-07, 2.1407e-03, 4.2644e-03, 5.0141e-03,
9.1932e-03, 5.1645e-03, 7.9613e-09, 1.3437e-02, 8.1442e-07,
6.0324e-03, 6.9080e-03, 1.8229e-03, 6.1963e-03, 7.1658e-03,
1.1932e-02, 1.1875e-03, 8.0651e-03, 1.1421e-02, 2.4946e-02,
4.1539e-03, 1.5138e-10, 2.0085e-13, 4.0516e-13, 2.9954e-02,
8.3709e-03, 3.6942e-03, 5.1183e-02, 5.2679e-13, 1.7226e-02,
3.6225e-03, 2.7673e-02, 8.0673e-03, 3.3364e-03, 3.3695e-02,
2.8535e-15, 4.6879e-03, 1.8904e-02, 1.5358e-02, 1.4966e-02,
1.3823e-02, 4.4650e-03, 8.0794e-13, 2.3731e-02, 3.8049e-03,
9.2273e-03, 4.6991e-03, 2.3623e-02, 2.9503e-03, 1.0473e-02,
3.7513e-02, 9.9128e-03, 2.1723e-02, 1.5941e-03, 4.5406e-02,
2.5284e-02, 8.8500e-15, 2.5378e-02, 5.4067e-03, 5.5883e-02,
2.3276e-02, 1.0463e-02, 2.1160e-02, 1.4845e-02, 4.7059e-02,
1.0868e-02, 1.2487e-02, 2.5815e-02, 1.4878e-02, 3.6792e-03,
1.0191e-02, 2.1322e-02, 7.4479e-03, 2.5276e-02, 1.7641e-09,
2.3546e-02, 5.8507e-03, 3.2746e-03, 3.0673e-02, 9.7895e-03,
2.6110e-11, 2.0301e-02, 2.4404e-02, 1.6003e-02, 2.5295e-03,
6.6987e-03, 5.1082e-03, 8.2862e-03, 1.5188e-03, 2.5924e-02,
9.5287e-04, 7.5685e-03, 1.6251e-02, 9.0352e-03, 6.7920e-03,
4.7632e-03, 5.1664e-09, 3.4988e-02, 3.0991e-02, 5.9015e-03,
3.0066e-03, 2.1890e-03, 2.2084e-02, 2.8663e-03, 1.5503e-02,
9.9153e-02, 2.6166e-02, 7.0402e-03, 1.3526e-02, 7.2020e-03,
1.8877e-12, 3.6811e-02, 1.0113e-03, 1.4059e-02, 1.1832e-02,
1.7269e-02, 2.3212e-02, 9.1356e-03, 5.2238e-03, 8.6943e-13,
8.8092e-03, 5.4724e-03, 3.3945e-02, 1.5528e-14, 1.0845e-02,
1.2344e-03, 4.0588e-03, 8.6161e-04, 2.5913e-02, 2.4235e-02,
6.5926e-03, 2.1641e-03, 4.1603e-03, 2.0683e-02, 9.0068e-03,
2.9246e-03, 2.3733e-12, 6.7850e-03, 1.8965e-02, 6.1481e-03,
5.7794e-11, 4.0696e-03, 5.4826e-03, 3.9190e-11, 3.6758e-02,
2.7583e-03, 1.7319e-03, 8.9602e-03, 9.4667e-03, 3.3400e-03,
3.4628e-03, 2.9662e-02, 1.3469e-02, 9.7367e-02, 1.0375e-02,
7.8521e-03, 3.0907e-12, 8.2250e-03, 2.4253e-02, 3.0203e-08,
2.7359e-03, 3.3542e-12, 7.4793e-14, 4.0583e-03, 1.5002e-05,
2.9705e-02, 5.1189e-03, 3.8250e-03, 8.0652e-02, 7.3081e-02,
2.8734e-03, 3.5167e-03, 7.8021e-04, 2.2556e-02, 4.9313e-02,
6.7593e-03, 3.8567e-04, 1.1143e-13, 1.4463e-02, 3.8817e-03,
6.2793e-03, 8.1507e-02, 1.0757e-14, 2.2426e-02, 1.4611e-02,
1.0395e-02, 1.9440e-02, 3.5411e-03, 1.1650e-02, 1.6200e-03,
7.0874e-03, 1.0921e-02, 7.8140e-03, 9.4227e-03, 5.7425e-03,
6.3261e-03, 1.5133e-10, 1.6517e-02, 2.6433e-02, 1.6696e-02,
2.3607e-03, 4.5254e-03, 1.4435e-02, 7.9935e-03, 3.7810e-02,
1.1055e-02, 1.0251e-02, 1.1947e-02, 1.3073e-15, 1.0083e-02,
1.0991e-02, 7.4614e-03, 2.6028e-03, 7.3397e-03, 5.4688e-03,
2.2999e-02, 1.9431e-02, 8.8905e-03, 1.0911e-02, 1.5871e-13,
9.4779e-03, 3.5101e-02], device='cuda:0')), ('layer2.1.conv1.weight', tensor([[[[ 2.9981e-02]],
[[ 6.2468e-03]],
[[-7.8216e-04]],
...,
[[-4.9786e-09]],
[[-9.1469e-03]],
[[ 5.1793e-02]]],
[[[-5.8691e-03]],
[[-6.9917e-03]],
[[-4.6123e-04]],
...,
[[ 3.6587e-09]],
[[ 2.9243e-03]],
[[-7.9964e-02]]],
[[[-5.3589e-02]],
[[-4.0799e-03]],
[[ 2.9705e-04]],
...,
[[-1.7033e-08]],
[[ 2.0213e-03]],
[[-3.2253e-02]]],
...,
[[[ 2.9383e-02]],
[[ 8.8350e-03]],
[[ 1.5486e-03]],
...,
[[ 5.4477e-08]],
[[ 3.4521e-03]],
[[-9.2993e-04]]],
[[[-4.2658e-02]],
[[-1.7373e-03]],
[[-3.9745e-04]],
...,
[[ 3.8903e-09]],
[[ 1.6505e-03]],
[[ 1.7884e-02]]],
[[[ 1.1154e-02]],
[[ 2.1871e-02]],
[[ 2.3255e-04]],
...,
[[ 2.5520e-08]],
[[ 7.0989e-04]],
[[ 2.7331e-02]]]], device='cuda:0')), ('layer2.1.bn1.weight', tensor([ 0.1143, 0.0757, 0.0914, 0.1120, 0.1156, 0.1359, 0.0838,
0.1183, 0.1306, 0.0885, 0.1084, 0.1306, 0.1053, 0.1468,
0.0738, 0.1036, 0.0873, 0.1163, 0.0768, 0.0728, 0.1531,
0.0943, 0.1124, 0.1238, 0.2047, 0.0920, 0.1435, 0.1362,
0.1109, 0.2188, 0.0865, 0.1191, 0.0889, 0.1052, 0.1366,
0.1328, 0.1251, 0.1528, 0.2020, 0.1533, 0.1039, 0.1004,
0.0843, 0.1067, 0.1523, 0.1989, 0.0936, 0.0802, 0.1393,
0.1333, 0.1425, 0.1023, 0.1807, 0.1137, 0.0921, 0.0990,
0.1294, 0.0921, 0.1386, 0.1249, 0.1101, 0.1374, 0.1408,
0.0873, 0.1088, 0.0506, 0.0885, 0.1001, 0.0911, 0.0805,
0.1002, 0.1057, 0.1325, 0.1071, 0.0992, 0.1108, 0.1346,
0.1271, 0.0677, 0.1194, 0.1223, 0.1385, 0.1079, 0.1060,
0.0611, 0.1348, 0.1082, 0.1542, 0.0887, 0.1154, 0.1260,
0.1264, 0.1798, 0.1189, 0.0979, 0.0981, 0.0791, 0.1435,
0.1183, 0.1051, 0.1222, 0.1360, 0.1385, 0.1431, 0.1288,
0.1498, 0.1085, 0.1627, 0.0939, 0.1143, 0.1391, 0.0636,
0.0967, 0.1255, 0.1151, 0.0964, 0.0787, 0.1066, 0.1409,
0.1014, 0.1143, 0.0863, 0.1536, 0.0985, 0.1438, 0.0895,
0.1469, 0.1272], device='cuda:0')), ('layer2.1.bn1.bias', tensor([ 0.0686, 0.1470, 0.0829, 0.0411, -0.0440, 0.0552, 0.1307,
0.0005, -0.0526, 0.2225, 0.2307, 0.3079, 0.0709, -0.0795,
0.1070, 0.0083, 0.0788, 0.0533, 0.0719, 0.0970, -0.0256,
0.1390, 0.0283, -0.0570, -0.0228, 0.1985, -0.0677, -0.0349,
0.0034, -0.1867, 0.0021, 0.0498, 0.1800, 0.0632, -0.0761,
0.2518, -0.0244, -0.0781, -0.1060, -0.0043, 0.0391, 0.0254,
0.0216, -0.0272, -0.0158, -0.1762, 0.1927, 0.1973, -0.0726,
-0.0281, -0.0334, 0.2211, -0.0630, -0.0096, -0.0309, 0.0570,
-0.0485, 0.1767, -0.0505, -0.0077, 0.0352, -0.0459, -0.0976,
0.0176, 0.1520, 0.0758, 0.1444, -0.0143, 0.2364, 0.0166,
0.0716, 0.0189, -0.0140, 0.1136, -0.0366, 0.0640, 0.1607,
-0.0204, 0.0636, 0.0286, 0.0240, -0.0004, 0.0181, 0.0243,
0.0977, -0.0513, 0.0864, -0.0497, 0.1234, 0.1664, -0.0528,
-0.0132, 0.3854, 0.1145, 0.0951, 0.1781, 0.0190, 0.0003,
0.0605, -0.0305, -0.0159, 0.0241, -0.0535, -0.0766, 0.3663,
-0.0140, 0.0631, -0.0458, 0.0234, 0.0123, -0.0590, 0.0549,
0.2084, 0.0515, 0.0630, -0.0188, 0.0648, 0.0512, -0.0948,
0.0446, 0.0229, 0.0669, -0.0550, 0.0246, -0.0370, 0.2319,
0.2566, -0.0807], device='cuda:0')), ('layer2.1.bn1.running_mean', tensor([ 0.2112, -0.0676, -0.0509, -0.1309, -0.0773, -0.0236, 0.1256,
0.0690, -0.0953, 0.1390, -0.1129, -0.1989, 0.1042, 0.0432,
-0.0644, -0.0731, 0.1379, -0.1678, -0.0048, 0.1069, -0.0132,
-0.0016, 0.0393, -0.0501, -0.1415, 0.0645, -0.0110, -0.0307,
-0.1156, -0.0643, -0.0048, -0.0282, 0.1375, 0.0580, -0.0239,
0.2228, -0.1323, 0.2020, -0.0121, -0.1251, -0.0199, 0.0439,
0.0892, 0.0714, 0.0833, -0.1727, -0.0613, -0.0880, 0.0614,
0.0610, 0.0153, -0.0945, 0.0143, -0.0434, -0.0100, 0.0320,
-0.0606, 0.1324, -0.0966, 0.0839, -0.0938, -0.1939, 0.0557,
-0.0120, 0.1363, 0.0139, 0.1667, 0.0507, -0.1094, -0.0415,
0.0887, -0.0161, 0.0323, 0.2570, -0.0149, 0.0574, 0.0790,
-0.0523, -0.0379, -0.0706, -0.0153, -0.0928, -0.0798, -0.1229,
0.0364, -0.0664, -0.0616, -0.0187, -0.0524, -0.0379, 0.0251,
0.1592, 0.2960, -0.0343, 0.0684, -0.0687, 0.0578, 0.0549,
0.1699, 0.0123, 0.0747, -0.0347, 0.0045, 0.0639, -0.3178,
0.0376, 0.0514, -0.0075, 0.0036, -0.1333, -0.2110, -0.0921,
-0.3032, 0.0471, -0.0526, 0.1479, -0.0448, 0.0144, 0.1332,
0.0900, -0.0410, 0.0059, 0.0290, -0.0104, -0.0039, -0.1429,
0.1303, 0.0266], device='cuda:0')), ('layer2.1.bn1.running_var', tensor([ 0.1117, 0.0407, 0.0683, 0.0203, 0.0142, 0.0404, 0.0877,
0.0464, 0.0109, 0.0696, 0.0374, 0.1664, 0.0624, 0.0299,
0.0261, 0.0187, 0.0278, 0.0214, 0.0163, 0.0307, 0.0178,
0.0769, 0.0235, 0.0113, 0.1069, 0.1250, 0.0123, 0.0063,
0.0315, 0.0446, 0.0154, 0.0436, 0.1175, 0.0646, 0.0209,
0.2712, 0.0212, 0.0148, 0.0342, 0.0349, 0.0404, 0.0243,
0.0080, 0.0150, 0.0148, 0.0192, 0.0706, 0.0513, 0.0112,
0.0283, 0.0190, 0.0472, 0.0203, 0.0090, 0.0150, 0.0461,
0.0343, 0.1256, 0.0293, 0.0269, 0.0261, 0.0223, 0.0258,
0.0147, 0.0334, 0.0164, 0.1091, 0.0212, 0.1229, 0.0146,
0.0514, 0.0228, 0.0228, 0.0429, 0.0085, 0.0349, 0.0354,
0.0517, 0.0197, 0.0434, 0.0321, 0.0649, 0.0196, 0.0174,
0.0223, 0.0161, 0.0266, 0.0176, 0.0329, 0.0919, 0.0114,
0.0179, 0.4486, 0.0211, 0.0596, 0.0996, 0.0191, 0.0271,
0.0589, 0.0021, 0.0116, 0.0448, 0.0423, 0.0229, 0.2804,
0.0157, 0.0710, 0.0146, 0.0230, 0.0360, 0.0216, 0.0265,
0.0539, 0.0819, 0.0363, 0.0102, 0.0485, 0.0139, 0.0336,
0.0288, 0.0167, 0.0198, 0.0153, 0.0415, 0.0201, 0.1139,
0.1710, 0.0127], device='cuda:0')), ('layer2.1.conv2.weight', tensor([[[[-2.6612e-02, -4.0164e-02, -1.7479e-02],
[-3.1620e-02, -5.3283e-03, -2.7553e-02],
[-2.4886e-02, -3.2047e-02, -2.3468e-02]],
[[-3.5126e-03, 3.2575e-03, -8.1418e-03],
[-7.0056e-03, 6.0254e-03, -1.0571e-02],
[-6.0508e-03, 7.2130e-04, -1.5162e-03]],
[[ 7.6138e-03, 1.8244e-02, 1.2507e-02],
[ 2.5248e-02, -3.7504e-02, 2.4530e-02],
[ 1.5743e-02, 2.3739e-02, 1.7725e-02]],
...,
[[ 4.3778e-03, 3.5878e-03, 7.4590e-04],
[ 4.4968e-03, -1.8396e-02, 3.6347e-05],
[ 5.4960e-03, 3.2917e-03, -2.2544e-03]],
[[-6.0618e-03, 1.2152e-02, 1.0627e-03],
[ 8.1596e-03, 6.4134e-03, 9.3486e-03],
[ 3.8096e-03, 1.8635e-02, 5.5144e-03]],
[[ 1.5398e-02, 1.1400e-02, 1.6990e-02],
[ 6.8940e-03, -2.7925e-02, 4.4624e-03],
[ 1.4584e-02, 9.9359e-03, 1.0279e-02]]],
[[[ 1.4447e-02, 1.0540e-02, 8.6184e-03],
[ 1.4049e-02, -5.1148e-02, 1.1761e-02],
[ 6.0427e-03, 1.3258e-02, 1.6813e-02]],
[[-9.7942e-04, 1.8536e-03, 1.2940e-04],
[ 4.5484e-03, 6.3970e-03, -5.7844e-03],
[-5.2112e-03, -9.8360e-03, -7.9912e-03]],
[[-6.5372e-03, -3.0737e-03, -6.6110e-04],
[-8.3917e-03, -4.3665e-03, -8.8227e-03],
[-7.9809e-03, -9.1811e-03, -9.4304e-03]],
...,
[[-3.0811e-03, 4.5446e-03, -2.2183e-03],
[ 1.3118e-02, -2.3592e-02, 1.2510e-02],
[-1.3352e-03, 1.6423e-02, 1.7204e-03]],
[[-3.9355e-03, -1.2400e-02, -5.3888e-03],
[-2.3812e-02, 6.2618e-02, -3.1402e-02],
[-1.5209e-02, -3.6339e-02, -2.3629e-02]],
[[ 1.1858e-02, 1.5748e-02, 5.0950e-03],
[ 8.5850e-03, -4.5303e-02, 5.7861e-03],
[ 2.5232e-03, 1.6618e-02, 1.0268e-03]]],
[[[ 1.1693e-04, -2.3821e-02, -1.0894e-02],
[-1.1472e-02, -3.5137e-03, -1.0219e-02],
[ 1.3313e-02, 3.7218e-02, 1.6308e-02]],
[[ 5.9613e-03, 2.5387e-02, 1.8554e-02],
[-1.4240e-02, -1.1972e-02, -1.0432e-02],
[-6.0892e-04, 5.4653e-03, 6.0121e-03]],
[[ 1.3706e-02, 1.8274e-02, 9.4376e-03],
[ 1.8834e-03, -6.9833e-03, -7.1998e-03],
[-1.0752e-02, -1.4026e-02, -1.2343e-02]],
...,
[[-3.5476e-02, -1.1280e-01, -5.9571e-02],
[-2.7227e-02, 8.1714e-03, -2.0615e-02],
[ 4.8816e-02, 1.4239e-01, 8.7312e-02]],
[[ 7.7823e-02, 2.1489e-01, 1.1521e-01],
[ 9.3477e-03, -2.4879e-02, -1.0115e-02],
[-4.7276e-02, -1.6804e-01, -1.0835e-01]],
[[-2.4170e-02, -4.6353e-02, -2.7198e-02],
[-1.6167e-02, 5.8170e-03, -1.1249e-02],
[ 1.6357e-02, 5.5004e-02, 3.2485e-02]]],
...,
[[[-1.6305e-02, -5.0020e-02, -1.5351e-02],
[-1.2441e-02, 5.3947e-02, -6.3890e-03],
[-2.4207e-03, 2.0496e-02, -3.4866e-03]],
[[ 3.4448e-03, 9.0776e-03, 3.1396e-03],
[-1.2736e-02, 2.3239e-02, -1.0691e-02],
[-1.9581e-02, 1.0343e-03, -1.1689e-02]],
[[ 2.0536e-02, 7.1731e-02, 4.3848e-02],
[-1.4356e-02, -6.6640e-02, -4.0626e-03],
[-1.6326e-02, -4.7985e-02, -1.0681e-02]],
...,
[[ 3.3409e-02, 8.4426e-02, 4.1283e-02],
[-1.0366e-02, -6.7267e-02, -8.4304e-03],
[-1.3965e-02, -3.6488e-02, -1.7230e-02]],
[[ 2.5791e-02, 3.6760e-02, 2.9285e-02],
[ 2.7356e-03, 3.1673e-02, 2.8453e-03],
[-1.0855e-02, 1.4385e-02, 5.6611e-04]],
[[ 1.0187e-02, 1.0588e-02, 1.7188e-02],
[-3.0925e-03, -3.9582e-02, 6.1720e-03],
[-5.8177e-03, -1.0946e-02, 6.8252e-04]]],
[[[-1.2185e-02, -1.3895e-02, -9.7944e-03],
[-1.6095e-02, 2.2727e-02, 7.0365e-04],
[-1.6365e-02, -6.8739e-03, -7.4789e-03]],
[[-6.2751e-03, -8.8369e-03, -1.0088e-02],
[ 4.7107e-03, 5.9371e-03, -7.3752e-04],
[ 1.7115e-03, 3.0230e-03, -8.1881e-03]],
[[-2.2295e-03, -2.0647e-03, -1.7058e-03],
[ 1.4152e-03, -1.7090e-02, -9.4503e-03],
[-2.1117e-03, -7.5780e-03, -7.4742e-03]],
...,
[[ 3.1252e-03, 1.0974e-02, -8.7916e-04],
[ 9.9409e-03, -2.5935e-02, 1.6085e-03],
[ 1.0177e-03, 3.6322e-03, 5.1371e-04]],
[[ 9.1652e-03, 4.5619e-03, -2.2218e-04],
[ 6.9943e-03, 1.6824e-02, -1.0487e-02],
[ 9.4374e-04, -1.5150e-02, -1.9611e-02]],
[[-1.6774e-03, 6.8204e-03, 3.9385e-03],
[-4.7882e-04, -3.7668e-03, 4.3544e-03],
[-1.0460e-03, 7.8029e-03, 3.3921e-03]]],
[[[ 1.0283e-02, 5.7515e-03, -1.3164e-02],
[ 1.0537e-02, -4.6643e-03, -2.3646e-02],
[ 1.8466e-02, -9.4791e-04, -1.6401e-02]],
[[ 8.7659e-03, -1.0505e-02, -4.4822e-03],
[ 1.2671e-04, 4.8271e-03, -1.0532e-02],
[ 3.8525e-03, -1.5733e-02, -9.0469e-03]],
[[ 3.0490e-04, -8.7700e-03, -2.2215e-03],
[ 1.9254e-02, 2.9673e-03, -1.2005e-02],
[ 1.3131e-03, -3.0526e-04, -9.1385e-03]],
...,
[[-1.6899e-03, -1.5691e-02, 1.6644e-02],
[ 1.1676e-02, -8.2631e-03, 2.4822e-02],
[-1.1040e-02, -1.7921e-02, 1.1375e-02]],
[[ 1.7830e-03, -1.2158e-02, 1.1729e-02],
[ 7.1818e-03, 6.8397e-04, 2.2245e-02],
[-1.5285e-03, -9.7472e-03, 1.3757e-02]],
[[ 9.1870e-03, 1.2355e-02, -4.5135e-03],
[ 8.5448e-03, -7.1429e-04, -2.3917e-02],
[ 3.4502e-03, 5.6828e-03, -1.2090e-02]]]], device='cuda:0')), ('layer2.1.bn2.weight', tensor([ 0.1776, 0.1253, 0.1823, 0.1996, 0.1886, 0.1409, 0.1549,
0.1246, 0.0863, 0.2138, 0.1544, 0.1554, 0.1665, 0.1341,
0.1169, 0.1823, 0.1384, 0.1631, 0.1488, 0.1989, 0.1235,
0.1653, 0.2177, 0.1262, 0.1818, 0.1841, 0.1535, 0.1297,
0.1605, 0.2361, 0.2205, 0.2062, 0.0967, 0.2203, 0.1388,
0.2097, 0.1593, 0.1837, 0.0797, 0.1574, 0.1405, 0.1758,
0.1517, 0.1396, 0.1525, 0.1645, 0.1626, 0.2054, 0.1696,
0.1582, 0.1589, 0.1655, 0.1751, 0.1953, 0.1255, 0.1181,
0.1275, 0.1586, 0.1367, 0.1509, 0.1663, 0.2498, 0.1959,
0.1912, 0.1517, 0.1599, 0.2867, 0.1220, 0.2525, 0.1890,
0.1916, 0.1373, 0.1461, 0.1639, 0.1563, 0.1762, 0.1324,
0.1051, 0.1793, 0.1718, 0.1759, 0.1436, 0.1498, 0.1712,
0.0930, 0.1331, 0.1355, 0.2007, 0.1051, 0.1681, 0.1448,
0.1426, 0.1590, 0.2092, 0.1544, 0.1967, 0.1247, 0.1491,
0.1739, 0.1149, 0.1266, 0.1627, 0.1850, 0.1815, 0.1988,
0.2299, 0.1590, 0.0869, 0.1557, 0.1388, 0.1757, 0.1473,
0.1639, 0.1859, 0.1761, 0.1901, 0.1914, 0.1486, 0.1887,
0.1612, 0.1703, 0.2223, 0.1447, 0.2026, 0.1558, 0.1597,
0.1068, 0.2155], device='cuda:0')), ('layer2.1.bn2.bias', tensor([-0.0184, 0.0471, 0.0683, 0.1018, -0.0076, 0.0603, 0.0339,
0.0681, 0.0400, -0.1420, -0.0781, -0.0290, 0.1200, 0.0270,
-0.0339, 0.0416, 0.0379, -0.1181, -0.0939, 0.0850, -0.0848,
0.1726, 0.0771, 0.0163, 0.0683, 0.0188, -0.0476, -0.0065,
-0.0019, -0.1350, -0.1825, 0.0421, 0.0739, 0.0721, -0.0071,
0.0286, -0.0353, 0.0735, -0.0379, 0.0404, 0.0831, 0.1374,
-0.0702, 0.1316, -0.0518, 0.0653, -0.0995, -0.0413, -0.1382,
0.1500, -0.1576, 0.1678, 0.0668, 0.0634, 0.0213, -0.0333,
0.0005, -0.0038, 0.0535, -0.0281, 0.0792, -0.1331, -0.1468,
-0.0365, 0.0778, 0.0568, -0.1759, 0.0385, -0.1446, 0.0476,
-0.1166, 0.0506, 0.0090, 0.0531, 0.0614, -0.0041, -0.0743,
0.0226, 0.0601, 0.0353, 0.0816, 0.0257, 0.0414, 0.0172,
-0.0382, -0.1142, 0.0041, 0.0522, 0.0693, 0.0740, -0.0474,
-0.0203, 0.0571, -0.1367, -0.0745, 0.0414, 0.0353, 0.0861,
-0.0493, -0.0339, 0.0198, -0.0254, -0.0596, 0.1878, 0.0945,
-0.1261, 0.0603, 0.0611, 0.0336, -0.0185, 0.1517, -0.0080,
-0.0808, -0.1123, -0.0406, -0.0137, 0.1976, -0.0426, 0.0726,
-0.0580, -0.0975, -0.1347, -0.0368, -0.0499, 0.0638, 0.0652,
0.0226, -0.0461], device='cuda:0')), ('layer2.1.bn2.running_mean', tensor([-0.0755, -0.1064, 0.1639, 0.1936, -0.2511, 0.0968, 0.1434,
0.0929, -0.1594, -0.0957, -0.0619, -0.0145, 0.0778, 0.0840,
-0.0511, 0.1027, -0.0194, -0.0469, -0.1445, 0.1702, 0.0112,
0.2628, 0.2508, 0.0444, 0.2848, 0.0941, -0.0587, -0.1810,
0.0054, -0.1493, -0.1787, 0.2605, -0.2624, 0.2600, -0.1839,
-0.2547, 0.0636, 0.1822, -0.1248, -0.1636, 0.0938, 0.3638,
-0.0910, 0.2369, -0.0288, 0.1791, -0.0545, -0.0578, -0.1302,
0.2971, -0.1782, 0.1815, 0.1359, 0.3126, -0.0154, -0.0642,
-0.0216, 0.0817, 0.1381, 0.0227, 0.1007, -0.1594, -0.1322,
-0.0740, 0.0807, 0.0979, -0.6248, 0.1634, -0.6697, 0.1824,
-0.0854, -0.1146, 0.0094, 0.0353, 0.1597, -0.0040, -0.0501,
0.0028, 0.1064, 0.2351, 0.3189, 0.1230, -0.0035, 0.0647,
-0.0700, 0.0802, -0.1744, 0.1175, 0.0117, 0.1496, -0.0238,
-0.0144, 0.2357, -0.0961, -0.1470, 0.3062, 0.0400, 0.2132,
0.0124, -0.0379, -0.0164, -0.0784, -0.1476, 0.3676, 0.3048,
-0.1588, 0.1181, -0.1313, 0.1071, 0.0424, 0.1955, -0.1524,
-0.0664, -0.4413, -0.1853, 0.0104, 0.4781, -0.0497, 0.3080,
-0.1262, -0.0456, -0.0722, -0.1703, -0.0155, 0.0969, 0.1538,
-0.0985, 0.0372], device='cuda:0')), ('layer2.1.bn2.running_var', tensor(1.00000e-02 *
[ 3.7423, 1.3799, 5.2627, 3.9147, 2.3844, 1.7165, 2.0942,
1.8393, 1.4561, 2.6326, 0.9871, 0.7858, 2.4364, 1.6992,
1.4911, 4.2959, 1.7580, 0.7789, 0.7001, 6.0610, 0.4787,
2.1353, 7.0459, 0.6329, 4.7184, 3.0803, 1.0986, 2.0989,
1.1170, 1.4898, 5.2644, 7.0888, 1.1159, 6.9768, 1.7665,
3.5994, 0.7126, 5.7401, 0.2321, 0.6907, 1.1065, 2.5608,
1.0433, 1.7183, 0.5999, 3.0878, 1.0653, 0.9611, 1.9434,
2.0628, 0.6152, 1.9821, 3.8698, 5.5445, 0.9709, 1.2993,
1.1869, 1.7536, 1.5575, 1.8916, 4.4341, 1.2942, 0.8997,
0.6319, 2.5562, 3.6593, 2.1778, 1.7454, 2.1016, 3.8873,
2.4259, 1.0912, 1.3430, 4.0809, 3.2527, 1.5779, 0.7843,
0.8562, 4.8620, 4.4895, 5.0816, 0.9927, 1.5393, 2.0523,
0.3082, 1.0425, 0.9380, 4.9724, 1.5268, 4.2497, 0.9216,
1.2957, 4.7747, 3.0762, 0.7316, 5.8499, 1.9204, 1.6922,
0.8858, 1.2042, 3.0828, 1.1886, 1.0838, 2.5486, 3.1880,
1.8077, 3.4598, 1.2683, 1.2793, 0.6135, 2.2290, 0.7540,
0.8391, 1.1255, 1.0946, 1.3901, 2.5971, 1.0777, 5.1527,
3.7159, 0.7759, 3.6072, 3.1282, 1.0585, 2.6542, 2.6445,
1.0523, 1.3533], device='cuda:0')), ('layer2.1.conv3.weight', tensor([[[[-4.4146e-02]],
[[ 2.0839e-02]],
[[-3.6888e-03]],
...,
[[-6.1172e-03]],
[[-1.4476e-02]],
[[ 2.2692e-02]]],
[[[ 6.6173e-04]],
[[ 2.4633e-04]],
[[ 4.5467e-03]],
...,
[[-1.3392e-03]],
[[ 5.3658e-03]],
[[-9.5005e-03]]],
[[[-2.4932e-02]],
[[-5.1053e-03]],
[[-1.6897e-01]],
...,
[[-1.2921e-04]],
[[ 4.2860e-04]],
[[ 9.3305e-03]]],
...,
[[[ 1.2192e-03]],
[[-3.8684e-02]],
[[-1.5341e-02]],
...,
[[ 4.2530e-02]],
[[ 4.9127e-02]],
[[ 9.1526e-02]]],
[[[ 4.1313e-04]],
[[-2.9892e-04]],
[[-1.2321e-02]],
...,
[[ 9.9619e-03]],
[[ 5.1484e-04]],
[[ 1.3669e-02]]],
[[[ 6.6747e-02]],
[[-7.0881e-03]],
[[ 4.9484e-03]],
...,
[[ 2.4755e-02]],
[[ 4.9751e-02]],
[[ 1.0807e-02]]]], device='cuda:0')), ('layer2.1.bn3.weight', tensor([ 1.3242e-01, -3.9867e-03, 2.3998e-01, 2.0384e-02, 1.4137e-01,
1.8754e-01, 1.0087e-03, 2.8137e-01, 1.3598e-01, 1.9805e-01,
2.3073e-01, 2.4595e-02, 8.5594e-02, 1.2895e-01, 2.5740e-01,
1.2544e-02, -1.3376e-03, 1.0062e-01, 1.0290e-01, 3.5137e-02,
9.0440e-04, 1.8675e-02, -2.8032e-03, 2.5723e-01, -5.0090e-03,
1.6177e-01, 1.7371e-01, 8.2233e-03, 2.3130e-01, 1.0041e-03,
1.4092e-02, 3.2567e-03, 2.4108e-01, 5.3649e-02, 6.5720e-02,
-8.4551e-05, -7.9860e-04, 5.1282e-03, -5.4507e-04, 8.2625e-03,
3.5324e-03, 8.1854e-03, 2.1627e-01, 1.1899e-01, 2.4078e-01,
1.1633e-02, 1.3839e-01, 5.9934e-03, -7.7509e-04, 5.1584e-02,
2.2356e-01, 8.2780e-05, -2.9549e-04, -2.1997e-02, 1.7908e-01,
-2.0951e-03, 1.7883e-02, 5.2630e-02, 1.6310e-03, 1.1751e-01,
2.3309e-01, 2.9326e-02, -7.3030e-04, 2.7353e-03, 3.9904e-02,
3.7391e-02, 3.6750e-02, 5.0586e-03, 3.3143e-05, -4.6347e-03,
1.7130e-02, -9.5308e-03, 3.3185e-02, 2.1624e-02, 4.4300e-03,
6.1887e-02, -3.2336e-02, -3.5629e-03, 7.5005e-02, 3.2207e-02,
2.4232e-02, 1.4763e-03, 2.6425e-01, 9.5805e-04, -4.7396e-03,
3.3679e-03, 2.6861e-02, 2.3871e-01, 2.2591e-01, 1.9161e-01,
2.5083e-01, 1.4579e-02, 1.3200e-02, 1.3942e-01, 2.5476e-01,
6.2012e-03, 2.5609e-03, 4.5695e-02, 2.8415e-02, 2.0069e-01,
7.2187e-02, 1.9426e-02, 1.9531e-01, -2.7378e-03, 1.8738e-01,
1.9436e-01, -2.4769e-03, 7.5864e-03, 6.1777e-04, 1.0982e-01,
-6.4235e-03, 5.4644e-03, 1.1123e-01, 5.3514e-02, 8.3084e-02,
8.9328e-03, -6.0282e-03, 1.6342e-02, -2.2930e-04, 5.4757e-02,
4.6124e-02, 5.9824e-02, 9.6306e-02, 8.2323e-03, 1.5161e-02,
2.6155e-01, 1.1137e-03, -1.9300e-02, 9.7766e-02, 1.0992e-02,
6.4315e-03, -9.2604e-03, 3.2524e-02, 1.5190e-01, 1.5307e-02,
8.5114e-02, -2.3247e-03, -7.6336e-03, 2.7737e-01, 4.9809e-02,
-3.6132e-02, 1.7670e-01, -1.0334e-02, 2.6541e-01, 1.7790e-01,
1.2973e-01, 1.2085e-01, 2.8980e-02, 6.1677e-03, 3.7799e-02,
3.7248e-03, 1.6790e-01, -2.9085e-03, 2.9735e-01, 1.7773e-01,
1.4563e-02, 1.4470e-01, 2.7867e-01, 6.5782e-03, 1.3230e-01,
1.9502e-02, 3.1345e-02, -1.8164e-03, 3.7131e-02, -4.8313e-03,
1.1612e-01, 3.3529e-02, 6.1328e-02, 7.6316e-04, 1.1034e-01,
2.3557e-01, 1.9592e-03, 8.2999e-03, 1.4917e-03, 1.2939e-01,
-4.3517e-03, 1.7499e-01, 2.9556e-03, 1.3388e-01, 3.5357e-02,
-5.1334e-02, 1.9123e-02, -8.4551e-03, 2.2817e-01, 2.4712e-02,
1.6226e-01, 2.7650e-01, 3.5893e-02, 1.9193e-01, 8.6777e-02,
2.5617e-02, 1.8742e-02, 7.9880e-02, 1.2931e-03, 2.8268e-01,
2.9270e-03, 2.5846e-01, -1.5192e-03, 2.4546e-01, 1.7163e-01,
1.7509e-06, 5.7166e-03, 5.7714e-03, 1.2100e-01, 2.7838e-01,
2.5758e-02, 1.2000e-01, 1.8500e-02, 7.8713e-03, -6.4103e-03,
2.6456e-02, -4.5954e-03, 8.0422e-03, 2.8520e-01, 2.3045e-01,
8.0622e-03, 8.1769e-03, -2.2261e-02, 2.9511e-01, -1.6241e-02,
1.5178e-01, 2.2163e-02, 2.5932e-01, 2.4696e-02, 4.4202e-03,
8.2374e-03, 6.2302e-03, 2.8438e-01, -6.9263e-03, -9.4133e-03,
8.9109e-03, -1.2923e-02, 2.7665e-01, 1.4762e-01, 2.7578e-01,
7.0166e-03, 2.8965e-01, 3.7330e-03, -3.8535e-03, 1.1965e-02,
2.5063e-01, 3.3793e-01, 1.5274e-02, 4.6105e-02, 1.1181e-02,
-4.0527e-04, 9.5739e-02, 1.2638e-02, 8.6265e-04, 1.5912e-01,
4.7056e-03, 5.2157e-02, -5.2267e-04, -1.4312e-02, -1.8230e-04,
1.5991e-01, -5.0568e-03, 5.3907e-02, -6.1297e-03, 5.7207e-03,
4.0776e-02, 5.8019e-03, 2.9346e-03, 1.3783e-01, -2.0062e-04,
3.0749e-02, 6.0224e-03, 7.1440e-02, 4.3121e-02, 3.3668e-02,
1.6093e-01, 3.8541e-02, 1.8803e-01, 1.4929e-01, 1.0741e-02,
2.2264e-01, 1.0222e-01, 1.0543e-01, 1.1332e-02, 2.3256e-04,
-7.7044e-03, 1.4923e-01, -3.1706e-03, 1.4994e-01, 2.1004e-01,
1.4621e-01, 7.5962e-03, -2.3061e-02, 1.8572e-02, -3.5922e-02,
1.2876e-01, 9.6549e-02, 6.5169e-03, 3.3788e-03, 2.3291e-03,
-5.7405e-03, 3.7034e-03, 2.5774e-01, 1.8605e-01, 2.5783e-05,
-2.5319e-02, 4.9742e-04, 1.2592e-01, 2.2677e-02, -8.8526e-03,
1.6606e-02, 3.4801e-02, 8.2358e-02, 2.8180e-02, 7.1734e-02,
-1.1104e-02, 2.1437e-01, 2.9919e-01, 1.8770e-01, 2.4306e-01,
2.7936e-01, -9.2975e-03, 1.7140e-01, 3.0950e-01, 1.6311e-01,
1.3238e-02, -1.6860e-02, 7.9346e-03, 6.2691e-02, 3.1991e-05,
3.4025e-01, 7.0419e-02, 2.1516e-02, 1.7150e-02, 2.3581e-01,
-8.3994e-03, 1.4497e-02, 3.3998e-01, 8.1498e-02, 1.3729e-02,
1.0834e-01, -8.9963e-04, 5.7597e-02, 7.8389e-03, 4.7418e-02,
1.7605e-01, 2.4862e-02, 2.1751e-01, 4.1433e-03, 2.3513e-01,
1.0048e-02, 2.5893e-01, 2.9535e-02, 2.1024e-02, -2.5114e-02,
-1.1503e-03, 9.1575e-02, -6.3913e-03, 1.2965e-01, 2.5463e-01,
1.1006e-02, 2.6259e-02, 4.0359e-02, 2.9381e-02, 1.2051e-03,
1.4009e-01, -1.6837e-02, 9.6909e-02, 6.2494e-02, 2.2921e-01,
-2.6400e-03, 9.0973e-03, -1.1027e-02, 1.7221e-02, -1.6165e-03,
2.6043e-01, 1.6975e-03, 4.6920e-04, 1.5349e-02, 3.5653e-03,
9.3198e-03, 5.3950e-02, 5.4591e-03, 4.0830e-02, 1.3520e-01,
9.0901e-03, -1.6302e-02, 1.6068e-01, -5.1534e-03, 9.0498e-03,
-3.4975e-03, 2.5242e-01, 1.3406e-01, 1.1536e-01, -3.4113e-03,
-1.5016e-03, -3.3296e-03, 1.1131e-01, 4.2994e-02, 3.2412e-02,
2.8762e-01, 1.4468e-01, 1.0058e-01, 8.6172e-04, 2.2942e-01,
3.3135e-01, 1.5762e-01, 4.4236e-03, 1.1463e-03, 1.2735e-01,
4.2306e-02, -2.2193e-02, 2.6118e-02, 9.4100e-02, 1.8414e-01,
1.0708e-02, -5.2921e-03, -2.3026e-03, 1.3227e-01, 2.6541e-01,
-5.8521e-03, 8.9120e-02, 3.1245e-03, -5.1983e-02, 9.6008e-02,
3.4769e-02, 3.0342e-01, 1.5691e-02, 1.4275e-02, 9.4966e-02,
5.5804e-02, 2.9243e-01, 2.6348e-01, 3.3957e-02, 1.2843e-02,
2.8020e-01, 3.3936e-02, 7.5655e-03, 2.8599e-01, -7.0982e-02,
4.6806e-04, 1.9623e-03, 1.3098e-02, 1.1181e-02, -3.9819e-03,
-4.8984e-03, 1.2586e-01, 1.1303e-01, 2.5781e-01, 6.4086e-02,
1.1150e-02, 2.5754e-01, 1.4200e-03, 2.2140e-03, 2.0586e-01,
3.9090e-01, -2.5558e-06, 2.3469e-01, 2.6226e-03, 2.6051e-01,
4.0117e-02, 1.1444e-03, 1.7378e-02, 2.4208e-01, 1.1203e-02,
5.3762e-03, -3.9318e-03, 2.7326e-02, -2.1225e-03, 2.2229e-01,
2.1667e-03, 1.7954e-02, 2.7006e-01, 5.7885e-02, -8.4252e-03,
2.1156e-02, 2.9087e-01, 2.9659e-01, 1.8163e-01, 2.8335e-03,
-9.4529e-03, -2.8763e-02, 1.0219e-01, 3.3065e-03, -4.2810e-04,
4.9658e-02, -2.2690e-03, 1.5966e-01, -3.5709e-03, 1.3090e-02,
1.1020e-01, 2.8386e-01, -8.1983e-03, 1.8239e-01, -2.9931e-02,
-4.2410e-03, 2.9046e-02, 1.7571e-01, 3.6939e-02, 1.0678e-01,
1.6147e-01, 1.0973e-02, 8.7048e-02, 2.5472e-01, -3.0390e-03,
1.5620e-01, 1.0042e-01, 2.5713e-03, 1.3052e-01, 6.2466e-03,
4.8908e-02, 4.2711e-02, 3.3092e-02, -1.1234e-02, 2.0025e-01,
-6.9633e-03, 1.8601e-01], device='cuda:0')), ('layer2.1.bn3.bias', tensor([-1.4303e-01, -4.1359e-02, 9.6345e-02, -1.7250e-03, -6.3559e-02,
2.4898e-02, 4.3319e-03, -1.1981e-01, -9.8809e-02, 7.8204e-02,
-6.3868e-02, -1.4853e-02, -2.2071e-02, -6.8391e-02, 1.3010e-02,
-3.5912e-03, 5.3247e-03, -6.2942e-03, -2.5711e-02, -2.9042e-02,
1.2719e-03, -3.5715e-02, 1.1996e-02, 5.5386e-02, 1.5934e-02,
-1.4835e-02, -8.4745e-02, 5.4219e-02, 3.1276e-02, 2.3295e-03,
-2.9171e-02, -7.6703e-02, 7.9058e-03, -1.7027e-02, -9.7281e-02,
-2.0484e-02, 1.1954e-02, 4.7291e-03, -7.3959e-03, 1.7732e-02,
-9.0220e-03, 8.9617e-07, -4.1675e-02, -9.9250e-02, -3.8707e-02,
-4.0530e-02, -1.3386e-01, -1.0893e-02, 1.1389e-03, -4.7458e-03,
-6.2041e-02, -3.0088e-03, 4.0795e-02, 1.8738e-03, -1.9352e-01,
-2.0363e-03, 3.3288e-02, -4.4849e-02, -3.5226e-03, 5.5271e-02,
-7.4215e-02, -2.4861e-02, -1.0789e-02, -1.0903e-01, -1.6468e-02,
-2.5583e-02, -1.5219e-01, 8.9196e-03, -3.7539e-04, 1.6317e-03,
1.2326e-03, -3.0860e-03, -2.7546e-02, -3.7632e-02, -8.8516e-03,
-3.0653e-02, -5.9028e-03, 6.1626e-04, -8.1051e-02, -4.3922e-02,
-9.5106e-03, -8.9856e-03, -3.1956e-02, -1.2650e-03, -1.0191e-02,
-2.3311e-03, 4.3025e-04, 2.7109e-02, -4.6026e-02, -2.3735e-02,
1.5982e-02, -1.6071e-02, -2.7669e-02, -9.0962e-02, 3.4019e-02,
-1.1714e-02, 9.0030e-03, -8.5057e-02, 2.6810e-03, 8.9295e-02,
-8.9216e-02, -3.3186e-02, -6.3577e-02, -7.6635e-04, -6.9414e-02,
1.0076e-01, 3.0637e-03, -1.4796e-02, 1.1292e-02, 4.2582e-02,
8.5550e-03, -6.4279e-03, 1.2091e-02, -4.9081e-02, -9.7279e-02,
-2.2004e-03, -1.2812e-03, -2.2187e-04, 1.0677e-02, -4.2782e-02,
-9.0975e-03, -3.4515e-02, -7.8852e-02, -8.6300e-03, -4.5601e-02,
-1.5004e-01, -2.5102e-03, -8.4100e-02, -1.5815e-02, -6.7395e-03,
-2.3238e-02, -1.0294e-02, -2.1451e-02, 1.2151e-01, -1.0588e-03,
-1.6892e-02, -4.8180e-03, -5.2033e-03, 1.4399e-02, -2.2758e-02,
-2.2461e-03, 3.0009e-03, -1.4381e-03, -9.5411e-03, 4.8236e-02,
-1.0254e-01, -1.7065e-01, -8.8658e-03, 4.8361e-04, -2.5649e-02,
-6.0476e-03, 9.2989e-05, -9.2383e-03, 7.3809e-04, -8.6246e-02,
-1.3256e-02, -1.5951e-01, -3.1092e-02, -3.8514e-03, -3.0750e-02,
1.5446e-02, -1.6930e-03, -7.0714e-04, -3.2160e-02, 1.0015e-02,
6.4493e-02, -5.2432e-02, -5.1374e-02, -1.8091e-02, 7.4934e-03,
-6.1742e-03, 3.8418e-03, 1.5771e-02, 8.2145e-03, 8.8184e-02,
-6.6368e-03, -8.0168e-02, -2.0362e-03, -2.9995e-02, -2.6609e-02,
-3.4397e-02, -1.7281e-02, 2.0384e-04, 7.9040e-03, -5.7181e-03,
-1.8650e-01, 6.7848e-02, -1.8846e-02, -1.3311e-01, 1.5576e-01,
-4.0909e-02, -7.0750e-03, -5.7762e-02, -2.7987e-03, -6.7647e-02,
5.7659e-04, 3.7205e-03, -1.0260e-02, -5.2115e-02, 7.8752e-02,
-8.8769e-06, -3.9048e-03, 6.7280e-03, -6.5355e-02, 8.2187e-02,
-3.9323e-02, 9.3245e-03, 1.8185e-02, 2.2181e-03, -1.6143e-02,
-1.8752e-02, 3.0577e-03, 2.4022e-03, 5.2618e-02, 2.9483e-02,
4.4747e-03, -2.6920e-02, -7.9864e-02, 3.2817e-02, 3.9993e-03,
-5.5510e-02, -1.6067e-02, 3.4400e-02, -2.6225e-02, 3.2143e-03,
-2.5083e-03, -3.1106e-03, -1.2949e-01, 3.0907e-03, -1.6687e-03,
-8.9696e-03, -9.7100e-04, 1.0100e-01, 1.8800e-03, 4.4217e-02,
1.9097e-02, 1.4490e-02, 8.7186e-03, 5.1226e-04, 1.0172e-03,
8.3062e-02, -1.4430e-01, -8.2419e-03, -1.0131e-01, -1.0135e-02,
-3.0764e-02, -1.2011e-01, -7.3087e-03, -1.4960e-02, 6.7472e-02,
1.0045e-02, -1.3479e-01, -2.1359e-03, -5.5752e-03, -6.7778e-03,
-1.4515e-01, 4.4448e-03, -1.4983e-02, 1.6710e-02, 1.2233e-02,
-3.9357e-02, 3.9256e-03, -9.1340e-03, -9.2751e-02, -1.7270e-03,
-1.2063e-01, 1.6213e-02, -1.2020e-01, -8.2898e-02, -8.8527e-02,
1.0714e-01, -4.5816e-02, 6.1812e-02, -7.3898e-02, 2.1483e-03,
6.2958e-02, -5.0742e-02, -4.7755e-04, -1.5459e-02, -2.2325e-02,
-2.1244e-03, -7.8404e-02, -1.0566e-02, -3.8241e-02, -3.6252e-02,
2.3692e-02, -1.3631e-02, -2.3484e-02, -2.4290e-02, -3.8339e-02,
1.0709e-01, 2.3524e-02, -2.4475e-03, -2.7759e-03, -4.4693e-02,
-4.1383e-03, 1.2485e-02, 5.5624e-02, -9.5404e-02, -1.8058e-03,
-1.2196e-02, -5.6406e-03, -1.1404e-02, -5.6079e-03, -2.2945e-03,
8.6154e-03, -4.6792e-02, 2.0591e-02, 1.5601e-02, -1.6393e-01,
-8.7655e-04, 8.3973e-02, 6.5980e-02, 1.9222e-02, -6.4712e-02,
-8.9406e-03, -1.7300e-02, -2.3836e-01, 3.7301e-02, -1.2117e-03,
8.7472e-03, -2.0330e-02, 1.0227e-01, -5.2300e-02, 1.1779e-02,
1.1245e-01, -9.6215e-02, 2.5785e-06, -4.3196e-03, -2.5858e-02,
8.9184e-03, -8.7743e-03, 4.6334e-02, -1.6391e-01, -1.7051e-02,
-7.1404e-02, -5.9554e-03, -1.7148e-01, -2.9559e-03, -1.9741e-02,
-1.0768e-01, -2.9664e-03, -1.6899e-01, -2.1230e-02, -6.6083e-02,
-1.2759e-02, 4.1966e-02, -4.0875e-02, 4.9974e-02, -1.2446e-01,
-5.5777e-03, 8.0188e-02, 3.4674e-03, -6.7782e-02, -1.1124e-01,
3.4792e-03, -1.7723e-02, -9.6419e-02, -7.1393e-03, -1.7892e-02,
-2.1882e-02, 1.2299e-02, -2.1861e-02, -1.2036e-01, -2.4058e-02,
1.1364e-02, 9.0925e-02, -1.7966e-02, 4.7963e-03, -9.1600e-04,
2.1050e-02, -1.0789e-01, 5.4525e-03, -6.9888e-03, 6.4915e-03,
-5.8863e-03, -4.2944e-02, -1.1018e-03, -1.8569e-02, -2.7626e-02,
-8.7634e-03, 2.8164e-03, -9.1081e-02, -4.3993e-03, -2.2357e-02,
-5.4756e-03, 1.7006e-03, -1.2821e-01, 1.0890e-01, -4.2571e-03,
-1.0445e-03, -5.0183e-03, -3.5273e-02, -6.6152e-02, -3.5577e-02,
-7.3482e-02, -1.6489e-01, -3.3737e-02, -8.6723e-03, -2.3769e-02,
4.3920e-02, 9.2937e-02, -1.9076e-05, 2.1228e-03, -1.6235e-01,
-5.3080e-02, -7.7377e-02, 1.5929e-02, -4.2757e-02, 3.0679e-02,
-1.1047e-02, -4.1060e-03, 2.6029e-03, -1.1071e-02, -4.5535e-02,
-1.4550e-02, -5.9604e-02, -1.3284e-02, -7.5299e-02, 3.6854e-04,
-1.2853e-02, -1.3467e-03, -1.3402e-02, -2.0708e-02, -6.1365e-02,
-1.0764e-02, -6.1896e-03, -3.0042e-03, -1.2082e-02, -1.0025e-03,
1.5994e-02, 4.5418e-02, 3.2360e-04, 1.0777e-02, -9.0291e-02,
-1.4739e-02, -1.9130e-02, -1.3679e-02, 1.2174e-02, 7.6363e-03,
-5.4397e-03, -1.5219e-01, -1.0369e-01, -1.1100e-01, 2.2357e-02,
2.5532e-02, 4.5459e-02, 5.4136e-03, 5.9550e-03, -3.8950e-02,
-8.2336e-02, -7.5211e-05, 7.1751e-02, 8.0743e-03, 2.7342e-02,
-4.9002e-02, 9.5575e-04, -7.5821e-03, -4.2720e-02, -1.5194e-03,
-5.9977e-03, 6.5210e-03, -2.7637e-02, -1.6451e-02, -1.8698e-01,
2.4471e-03, -2.9155e-02, 4.5423e-02, -7.8423e-02, -8.4752e-03,
-3.4671e-02, -1.0741e-01, 5.1450e-02, -1.2092e-01, -4.6671e-02,
-2.7426e-03, -5.7088e-02, -8.7176e-02, 2.2790e-02, -1.2400e-02,
-3.1559e-02, 5.8307e-03, 4.0074e-02, -8.5271e-03, 3.3860e-04,
-7.4499e-02, 6.0329e-03, 7.9079e-04, -5.2139e-02, -2.6035e-02,
-4.8131e-03, 1.4255e-01, -5.6571e-02, 1.6214e-02, -1.8812e-01,
-1.4541e-01, -7.4016e-03, 3.8062e-02, -1.9768e-02, 3.5405e-04,
-7.8156e-02, 7.2524e-02, -1.7100e-02, -1.9132e-02, -1.6783e-03,
-1.7947e-01, -1.2886e-01, -6.4873e-02, -1.9259e-02, 9.7423e-02,
5.2216e-03, -1.3475e-01], device='cuda:0')), ('layer2.1.bn3.running_mean', tensor([-7.3104e-02, -6.5895e-03, -5.2702e-02, -5.4670e-03, -2.8311e-02,
-5.6996e-02, 5.7032e-03, -4.2091e-02, -3.6394e-02, -7.2781e-02,
6.9172e-02, -2.0221e-02, -3.3049e-02, 4.2856e-03, 3.9047e-03,
-1.2676e-02, 1.2415e-02, 6.2415e-03, 5.6992e-03, 1.0380e-02,
1.1472e-03, -8.5680e-03, 4.9710e-03, -5.4391e-02, -6.5014e-03,
-2.3932e-03, 1.2406e-01, 1.1840e-03, -2.1098e-02, -9.5961e-03,
-1.9557e-03, 1.2762e-03, -2.9295e-02, -6.6804e-03, 1.2109e-02,
-1.5353e-02, 7.5480e-03, -3.3568e-03, -4.9770e-04, -3.4601e-02,
7.7038e-03, 4.7085e-03, 6.0444e-03, 6.7987e-02, -4.6638e-03,
-3.3085e-03, -1.3536e-02, -6.7086e-03, -2.9475e-03, -3.6737e-02,
-3.4949e-02, -2.0757e-04, 1.2485e-02, -3.9728e-03, 3.8693e-02,
1.0591e-04, 2.5999e-02, -1.0409e-02, -1.2832e-02, -5.5322e-02,
-3.2871e-02, 2.0141e-02, -7.8598e-03, 2.6851e-03, -4.8583e-03,
-4.8565e-03, -9.5214e-03, 8.5237e-03, -1.4816e-04, -1.5832e-03,
3.0344e-02, 5.3059e-03, -1.6371e-02, -9.8961e-03, 6.1859e-03,
-6.9488e-02, -3.2659e-04, 3.5357e-03, 7.4359e-03, 1.0181e-02,
-9.5350e-05, -7.9336e-04, -2.1882e-02, 7.5678e-03, 3.5933e-03,
-1.9701e-03, -7.1269e-03, -1.4421e-02, -1.4857e-02, -2.3871e-02,
-6.0725e-02, 3.8598e-03, -9.1335e-03, -1.4255e-02, -6.3984e-02,
3.2735e-04, -2.7524e-03, 3.8393e-02, 1.5865e-02, 4.7508e-02,
2.1326e-02, -1.6452e-02, -9.8845e-03, -1.3708e-03, -5.7388e-02,
-8.1981e-02, 3.5285e-03, -4.7741e-03, 8.8903e-03, -2.4387e-02,
-5.5606e-03, -7.7777e-03, -4.4169e-02, 5.5865e-03, 2.5491e-02,
-8.6096e-03, 6.0065e-03, 7.2288e-03, -3.2082e-03, -1.2790e-02,
1.2509e-03, 4.6565e-03, 5.7995e-03, 3.4350e-03, -8.1138e-03,
-2.9890e-02, -1.1385e-03, 4.4516e-04, -4.1244e-02, -1.1517e-02,
-2.3265e-03, -1.9710e-03, -1.7366e-02, -1.6366e-01, 4.1451e-04,
-2.4466e-02, -1.1105e-02, 1.7795e-02, -6.6430e-02, 1.4071e-02,
-1.0679e-02, -8.1140e-03, -5.9057e-04, 2.1331e-03, -7.2365e-02,
-2.8459e-02, -7.4908e-02, 1.1751e-04, 1.6431e-04, -1.7645e-03,
-3.2895e-03, 1.6433e-03, 7.4341e-03, -3.6542e-02, -3.6851e-02,
5.1665e-03, 8.9338e-03, -4.6003e-02, 1.8433e-02, -2.9550e-02,
3.1063e-04, 6.8003e-03, -5.7610e-03, 1.8504e-02, 3.4253e-03,
-1.1947e-02, -3.0903e-02, -4.2841e-03, 9.6860e-03, -2.1825e-02,
-4.9221e-02, -1.2772e-02, -5.2598e-03, -4.8745e-03, -3.8757e-02,
5.6482e-03, -5.2290e-02, -1.6862e-03, -3.5410e-02, -4.3679e-03,
-5.7273e-03, 1.4717e-03, -3.1202e-05, -1.1698e-01, 2.9406e-03,
-3.5349e-02, -5.2577e-02, 3.3885e-03, 6.3404e-02, 4.4941e-02,
1.5627e-03, 1.3940e-02, -1.2106e-02, -2.3574e-02, -6.3787e-02,
-1.4999e-03, -3.2744e-02, -2.6092e-03, 2.7462e-02, -6.8087e-02,
-5.3801e-07, 6.3947e-03, -4.4001e-04, -1.2159e-02, -9.5738e-02,
-1.2952e-02, -1.0871e-01, 4.5878e-04, -1.3732e-02, 2.4854e-03,
-7.3948e-03, -3.5477e-04, 9.5756e-03, -7.8426e-02, 1.5429e-02,
6.9204e-03, -1.8197e-02, -2.3305e-02, -7.8677e-02, 2.4066e-03,
4.7500e-02, -1.1916e-03, -5.5488e-02, -3.2571e-02, -4.8879e-03,
6.5105e-03, -3.0316e-03, -8.7397e-02, 1.5615e-03, -2.0961e-03,
-6.3610e-03, -1.2748e-02, -9.4821e-02, -2.4581e-02, -6.9738e-02,
5.7550e-03, 3.8107e-02, -1.4101e-02, -2.9115e-03, -1.1620e-02,
-2.7054e-02, -3.8226e-03, -3.1116e-03, 9.9870e-03, -9.7359e-03,
5.9056e-04, 5.9077e-02, 2.0651e-02, -3.9846e-03, -6.3973e-03,
-6.6491e-03, -1.5569e-02, -3.1777e-04, 6.6336e-03, -8.4751e-03,
-1.2957e-02, -2.3958e-03, 3.4020e-03, -6.0984e-04, -1.4305e-02,
-6.5529e-03, 3.3336e-04, -5.4599e-04, -3.9539e-02, -4.7352e-03,
2.7413e-03, 3.2302e-03, 6.1794e-02, -3.8759e-03, 2.3865e-02,
-7.5638e-02, -6.7079e-03, -4.0841e-02, -1.6707e-02, -2.8704e-04,
-4.9017e-02, -1.1377e-02, 1.4460e-03, 1.0323e-03, 1.5356e-02,
-1.5433e-03, -6.8280e-03, -1.5277e-03, 1.4742e-02, -4.6223e-02,
1.3642e-02, 1.3723e-02, 3.3458e-03, 7.5365e-03, -1.7975e-02,
3.8394e-02, -3.3724e-02, -4.8983e-03, 4.6362e-03, 1.0390e-03,
6.3180e-03, 9.2550e-03, -4.5881e-02, -4.1666e-02, 3.9789e-04,
-1.1633e-02, 2.4241e-03, -6.3924e-02, 1.4917e-03, 1.2529e-02,
1.0604e-02, -8.5949e-03, 1.4424e-02, -6.6100e-03, 6.9325e-03,
-1.4369e-02, -1.1850e-01, -5.7397e-02, -5.1382e-02, -8.0655e-02,
-5.9557e-02, 1.0300e-02, 5.9866e-04, -7.8677e-02, -2.2457e-02,
-4.8635e-03, 5.3699e-03, 5.5923e-03, -1.1712e-02, -7.1570e-03,
-7.0871e-02, 1.3285e-02, 4.2864e-03, -5.0477e-03, -5.6924e-02,
-1.5832e-02, -1.9788e-04, -6.2556e-02, 2.8124e-02, -7.9552e-03,
-2.5157e-02, -3.7078e-03, 1.1330e-02, 6.7233e-04, 8.9842e-03,
-6.7043e-03, -3.0387e-03, -4.9099e-02, -7.0104e-03, -3.3052e-02,
-9.4303e-03, -5.4229e-02, 7.9485e-03, -3.3969e-03, -1.9796e-02,
-3.1538e-03, 2.7441e-02, 1.7547e-02, -1.8895e-02, -2.0315e-02,
-9.0816e-04, 3.4397e-03, 1.2040e-02, -1.9383e-02, -1.0360e-02,
4.3151e-03, -1.8167e-02, -2.1722e-02, -3.4029e-02, -1.9369e-02,
2.6532e-03, -1.7270e-03, -5.0894e-03, 1.5491e-02, 1.7590e-03,
1.4795e-02, 2.6001e-04, 3.3926e-03, -6.6253e-03, 3.3196e-03,
1.1484e-02, 2.3542e-03, -9.5027e-04, -2.3376e-02, -3.8820e-02,
5.9331e-03, -5.2390e-03, 8.2183e-02, 6.2139e-03, 5.0152e-03,
-1.5765e-03, -7.8472e-02, -4.7523e-02, -5.9338e-03, -5.0659e-03,
-2.0892e-03, 1.0787e-03, -4.2671e-02, -1.9614e-02, -1.3650e-02,
-8.0085e-03, -6.4105e-03, -9.1372e-02, -3.6707e-03, -3.7078e-02,
-5.7575e-02, 3.5966e-03, -4.3302e-03, 1.9795e-03, 7.8777e-02,
-7.7683e-03, -6.3418e-03, -9.0245e-03, 1.5449e-02, -4.3861e-02,
2.5633e-03, -1.2398e-02, 2.7364e-03, -5.0364e-02, -1.1859e-02,
9.6033e-03, -3.5791e-03, 1.0714e-02, 2.9078e-02, -1.7720e-02,
4.1482e-03, 1.3466e-02, -4.6495e-03, -5.5041e-03, -6.5482e-03,
-8.0552e-03, -2.7858e-02, -4.6387e-02, -4.9681e-03, -8.5905e-03,
-2.0760e-02, -1.8202e-02, -1.8878e-03, -2.9344e-02, -3.4749e-02,
4.4421e-04, -1.2832e-03, 1.5593e-02, -5.2080e-03, 2.9491e-03,
-6.5308e-03, -6.7059e-02, -8.3318e-03, -4.0129e-02, 5.9158e-02,
-4.5696e-03, -5.7688e-02, -4.9331e-03, -9.7980e-03, 7.5509e-03,
-2.1565e-02, -7.9175e-05, -2.0545e-01, -3.2864e-04, -3.9596e-02,
-3.8006e-02, -6.3206e-03, 8.0633e-03, -7.4766e-02, 5.9139e-03,
1.5363e-02, -1.4610e-04, -3.0435e-03, 2.2991e-03, -4.0825e-02,
-1.1877e-02, -1.7128e-02, -5.1054e-02, -3.6603e-02, -7.8968e-04,
2.6431e-02, -4.4726e-02, -5.8300e-02, -1.2016e-02, 7.1723e-03,
-3.7248e-03, -1.8516e-02, -1.7323e-02, 5.8025e-03, 8.3483e-03,
-1.5714e-03, 1.8471e-02, -1.5959e-01, 1.2365e-02, 1.4373e-03,
1.1710e-02, -3.3287e-02, 1.9114e-03, -2.8794e-02, -3.7549e-02,
-4.1518e-03, 1.2998e-02, -4.4517e-02, -2.6236e-02, -7.1854e-02,
-1.4890e-01, -3.5874e-03, -8.2806e-03, -6.2501e-02, -9.3224e-03,
-4.7919e-02, 1.5431e-03, 7.7715e-03, -7.5346e-03, -6.9781e-03,
-8.0175e-03, -4.3606e-03, -7.7353e-03, 1.9256e-04, -2.2291e-02,
1.0368e-02, -2.1899e-02], device='cuda:0')), ('layer2.1.bn3.running_var', tensor([ 1.8065e-03, 1.7684e-04, 5.7473e-03, 2.8833e-04, 2.0792e-03,
2.8216e-03, 2.7691e-04, 5.4520e-03, 1.3991e-03, 3.2460e-03,
2.5605e-03, 2.9506e-04, 1.6223e-03, 1.7437e-03, 4.2353e-03,
1.6818e-04, 4.4357e-04, 1.0446e-03, 1.2015e-03, 5.5684e-04,
1.4779e-04, 1.6888e-04, 2.0297e-04, 9.7422e-03, 2.8524e-04,
2.9071e-03, 2.7675e-03, 3.2873e-04, 5.7709e-03, 2.5921e-04,
1.5948e-04, 2.4067e-04, 7.5891e-03, 5.4637e-04, 5.3220e-04,
2.0015e-04, 7.6052e-04, 2.3808e-04, 4.1868e-06, 7.7141e-04,
1.8883e-04, 1.6415e-04, 3.7263e-03, 1.1569e-03, 4.7514e-03,
1.7265e-04, 1.1907e-03, 1.2395e-04, 2.2291e-04, 6.8749e-04,
4.5630e-03, 1.6458e-07, 2.9088e-04, 2.9325e-04, 3.1031e-03,
1.2494e-04, 4.1107e-04, 5.1929e-04, 3.8264e-04, 2.5307e-03,
2.8554e-03, 4.2232e-04, 1.0275e-04, 1.0462e-04, 3.0308e-04,
3.6306e-04, 1.6136e-04, 1.9083e-04, 4.7017e-08, 1.1058e-04,
2.6053e-04, 2.4533e-04, 5.0102e-04, 3.5102e-04, 2.9600e-04,
4.6931e-04, 3.7439e-04, 3.1511e-04, 2.4627e-03, 1.4726e-04,
3.5177e-04, 1.5882e-04, 4.5895e-03, 1.5756e-04, 1.3311e-04,
3.0983e-04, 5.2442e-04, 7.2623e-03, 3.5807e-03, 2.4360e-03,
7.0770e-03, 2.1287e-04, 1.3758e-04, 8.1683e-04, 1.0127e-02,
1.0287e-04, 2.6667e-04, 9.4967e-04, 3.1092e-04, 4.2489e-03,
6.5016e-04, 1.8356e-04, 2.9720e-03, 2.5564e-04, 2.5009e-03,
3.8302e-03, 1.4539e-04, 1.1717e-04, 3.2719e-04, 2.0535e-03,
2.4762e-04, 2.9280e-04, 2.2021e-03, 7.7282e-04, 8.9405e-04,
4.6988e-04, 1.3394e-04, 2.5785e-04, 1.9401e-04, 8.4309e-04,
5.5660e-04, 5.4348e-04, 1.6092e-03, 1.5144e-04, 1.7207e-04,
5.6587e-03, 2.8012e-04, 1.7776e-04, 1.2789e-03, 1.8843e-04,
2.0253e-04, 2.7778e-04, 2.6398e-04, 2.5298e-03, 5.0912e-04,
1.3234e-03, 2.0595e-04, 2.5482e-04, 9.8349e-03, 7.8488e-04,
4.2872e-04, 4.2951e-03, 1.6461e-04, 5.4741e-03, 2.0527e-03,
8.3424e-04, 9.9664e-04, 3.2884e-04, 2.2876e-04, 7.2095e-04,
2.6652e-04, 4.5099e-03, 3.8767e-04, 5.6442e-03, 1.9498e-03,
4.3829e-04, 1.4251e-03, 8.0431e-03, 2.3540e-04, 1.8366e-03,
3.0857e-04, 5.8510e-04, 3.1959e-04, 4.3504e-04, 3.2166e-04,
1.2592e-03, 3.0991e-04, 6.6878e-04, 1.1557e-04, 1.3872e-03,
6.2586e-03, 6.8718e-04, 8.1113e-04, 3.6898e-04, 2.7942e-03,
2.0749e-04, 2.1834e-03, 3.3546e-04, 1.7794e-03, 2.6544e-04,
3.5356e-04, 3.5087e-04, 1.7991e-04, 3.0848e-03, 2.8921e-04,
2.6099e-03, 1.0786e-02, 6.2849e-04, 2.9858e-03, 2.0358e-03,
9.3294e-04, 1.4565e-04, 7.3638e-04, 3.5475e-04, 8.3664e-03,
1.9251e-04, 5.5038e-03, 1.4877e-04, 4.9255e-03, 3.0016e-03,
6.2146e-13, 3.2148e-04, 1.8592e-04, 2.5337e-03, 9.9726e-03,
1.4493e-04, 1.5415e-03, 3.3260e-04, 8.3619e-04, 7.7059e-06,
4.2944e-04, 3.5512e-04, 5.9856e-04, 9.7215e-03, 3.5777e-03,
1.5611e-04, 3.5399e-04, 3.6627e-04, 1.0201e-02, 2.1767e-04,
1.5395e-03, 2.0830e-04, 5.2813e-03, 1.7795e-04, 2.1954e-04,
2.7584e-04, 1.9749e-04, 6.3393e-03, 1.5825e-04, 3.4096e-04,
1.8415e-04, 2.2748e-04, 6.5050e-03, 2.0448e-03, 1.0793e-02,
1.7957e-04, 6.3697e-03, 3.0434e-04, 3.4699e-04, 2.1597e-04,
8.3008e-03, 7.3781e-03, 2.1809e-04, 5.4077e-04, 1.0952e-04,
8.3255e-05, 1.2575e-03, 3.8761e-04, 1.1788e-04, 2.7861e-03,
2.4554e-04, 1.5984e-03, 1.5830e-04, 2.3909e-04, 1.5100e-04,
1.2527e-03, 1.4163e-04, 7.8373e-04, 3.5287e-04, 4.2882e-04,
3.8358e-04, 2.4814e-04, 2.0979e-04, 3.0290e-03, 2.3285e-04,
3.3687e-04, 1.7386e-04, 5.3231e-04, 8.9849e-04, 3.0025e-04,
4.4275e-03, 2.2273e-04, 5.7924e-03, 4.6605e-03, 2.3980e-04,
3.7408e-03, 1.5968e-03, 1.2087e-03, 1.6601e-04, 1.5034e-04,
1.7814e-04, 1.6670e-03, 1.7044e-04, 1.5804e-03, 3.1047e-03,
2.9027e-03, 2.9255e-04, 2.6914e-04, 2.6970e-04, 1.7756e-04,
2.3868e-03, 1.1714e-03, 1.2349e-04, 2.1334e-04, 8.4690e-05,
2.6821e-04, 2.5417e-04, 7.9179e-03, 1.1981e-03, 1.7162e-07,
3.5500e-04, 8.1284e-05, 1.5109e-03, 3.0392e-04, 1.5505e-04,
3.1130e-04, 1.7574e-04, 1.2650e-03, 3.0875e-04, 3.7677e-04,
1.8110e-04, 4.2139e-03, 1.0335e-02, 4.0699e-03, 2.6176e-03,
7.3961e-03, 1.2333e-04, 2.1510e-03, 8.8504e-03, 2.6868e-03,
1.6790e-04, 3.0408e-04, 3.7876e-04, 7.7072e-04, 2.8416e-04,
1.5570e-02, 4.7174e-04, 2.6802e-04, 1.8269e-04, 7.3000e-03,
4.7970e-04, 2.3803e-04, 1.1697e-02, 4.0009e-04, 1.7045e-04,
1.0604e-03, 1.9031e-04, 5.2174e-04, 2.1644e-04, 6.3001e-04,
3.3444e-03, 2.2933e-04, 3.4998e-03, 1.5643e-04, 4.1936e-03,
4.9057e-04, 1.0012e-02, 3.2749e-04, 3.1486e-04, 5.9723e-04,
2.9420e-04, 1.7358e-03, 3.6815e-04, 1.4951e-03, 4.8119e-03,
1.9728e-04, 3.5276e-04, 5.4712e-04, 2.5555e-04, 2.0094e-04,
1.5670e-03, 3.8915e-04, 1.5269e-03, 2.0591e-04, 4.1069e-03,
3.7247e-04, 3.3099e-04, 2.2075e-04, 3.7931e-04, 2.1037e-04,
6.8198e-03, 1.8570e-04, 3.4420e-04, 3.2069e-04, 4.5261e-04,
2.0338e-04, 3.9312e-04, 1.6715e-04, 3.5295e-04, 1.7274e-03,
2.6692e-04, 2.0203e-04, 2.6084e-03, 1.5091e-04, 2.2793e-04,
1.5913e-04, 6.6154e-03, 2.7887e-03, 2.8185e-03, 1.1635e-04,
1.7640e-04, 1.8651e-04, 1.3541e-03, 1.0794e-04, 1.1734e-03,
6.7787e-03, 1.8612e-03, 9.3033e-04, 1.9511e-04, 5.8949e-03,
1.3303e-02, 3.7326e-03, 2.7949e-04, 1.3302e-04, 1.2269e-03,
6.0827e-04, 2.6532e-04, 2.8417e-04, 1.1097e-03, 2.9420e-03,
2.3460e-04, 1.9668e-04, 2.5483e-04, 2.4877e-03, 7.4996e-03,
1.1660e-04, 5.4543e-04, 1.9032e-04, 3.2431e-04, 4.0127e-03,
2.1391e-04, 7.5656e-03, 1.4229e-04, 1.9224e-04, 4.7521e-04,
9.8862e-04, 6.4117e-03, 7.8934e-03, 6.1788e-04, 1.6996e-04,
4.7662e-03, 4.9845e-04, 3.1713e-04, 4.5554e-03, 4.9214e-04,
1.4389e-04, 1.4287e-04, 3.0449e-04, 1.8864e-04, 1.6783e-04,
1.3834e-04, 1.6747e-03, 7.7575e-04, 6.4431e-03, 8.6821e-04,
2.7605e-04, 6.8137e-03, 2.8898e-04, 3.1493e-04, 2.5065e-03,
1.1102e-02, 1.9931e-09, 2.7873e-03, 2.0447e-04, 4.8359e-03,
6.3681e-04, 2.5338e-04, 1.5706e-04, 4.8163e-03, 5.6663e-04,
1.6117e-04, 2.2566e-04, 9.3337e-05, 2.0569e-04, 3.8339e-03,
2.2080e-04, 7.8370e-05, 9.4149e-03, 6.4084e-04, 1.6369e-04,
2.0013e-04, 6.2440e-03, 1.2931e-02, 1.5951e-03, 1.6392e-04,
1.8718e-04, 2.9011e-04, 3.6467e-04, 2.5782e-04, 2.0552e-04,
5.6957e-04, 2.2749e-04, 2.7586e-03, 1.6868e-04, 3.1368e-04,
9.5154e-04, 6.4231e-03, 2.7392e-04, 2.0126e-03, 4.4596e-04,
1.3491e-04, 6.3011e-04, 3.1016e-03, 6.1103e-04, 1.4851e-03,
1.1205e-03, 2.0812e-04, 1.6134e-03, 7.7553e-03, 2.5126e-04,
1.4333e-03, 1.1610e-03, 1.6989e-04, 2.0333e-03, 1.5638e-04,
5.5225e-04, 2.0417e-04, 5.1328e-04, 2.2321e-04, 3.9222e-03,
3.7744e-04, 4.5222e-03], device='cuda:0')), ('layer2.2.conv1.weight', tensor([[[[ 7.4153e-03]],
[[ 6.2506e-03]],
[[-2.0115e-02]],
...,
[[ 1.9778e-02]],
[[ 4.9028e-03]],
[[ 2.3533e-02]]],
[[[ 3.6302e-04]],
[[-3.4109e-02]],
[[-1.8630e-02]],
...,
[[ 8.8631e-03]],
[[ 1.2520e-02]],
[[-2.4100e-02]]],
[[[-3.1197e-03]],
[[-2.0085e-02]],
[[ 1.7950e-02]],
...,
[[ 2.2827e-02]],
[[-4.4781e-02]],
[[ 8.4873e-04]]],
...,
[[[ 5.7927e-03]],
[[-4.4353e-04]],
[[-1.4330e-03]],
...,
[[ 3.3877e-02]],
[[ 8.9764e-03]],
[[ 5.4745e-02]]],
[[[ 1.2431e-02]],
[[ 3.7703e-03]],
[[-3.3974e-03]],
...,
[[ 5.1701e-02]],
[[ 3.8852e-02]],
[[-7.2459e-03]]],
[[[ 2.5396e-02]],
[[ 1.4263e-03]],
[[ 5.8049e-02]],
...,
[[-7.5972e-03]],
[[ 1.3941e-02]],
[[-2.7615e-02]]]], device='cuda:0')), ('layer2.2.bn1.weight', tensor([ 0.1067, 0.1330, 0.1966, 0.2041, 0.1228, 0.1568, 0.1682,
0.1349, 0.2218, 0.1687, 0.1892, 0.1968, 0.1744, 0.1540,
0.2011, 0.1394, 0.1831, 0.1286, 0.1294, 0.1838, 0.1811,
0.1735, 0.1889, 0.2164, 0.1570, 0.1866, 0.1985, 0.1465,
0.1640, 0.1273, 0.2273, 0.1822, 0.1239, 0.1486, 0.1905,
0.1189, 0.1204, 0.1201, 0.2163, 0.1234, 0.1987, 0.2345,
0.1419, 0.1329, 0.2138, 0.1867, 0.1982, 0.1370, 0.2064,
0.1887, 0.1521, 0.1634, 0.1765, 0.1685, 0.1898, 0.2198,
0.1706, 0.1577, 0.1609, 0.1754, 0.1328, 0.1600, 0.1973,
0.1794, 0.1991, 0.1767, 0.1928, 0.1742, 0.1744, 0.1367,
0.2161, 0.2227, 0.1319, 0.1864, 0.2205, 0.1355, 0.2003,
0.1855, 0.1881, 0.1726, 0.2034, 0.1853, 0.1983, 0.2177,
0.1946, 0.1802, 0.1104, 0.2102, 0.1276, 0.1824, 0.1731,
0.1329, 0.1180, 0.1595, 0.1543, 0.1727, 0.1970, 0.1179,
0.1991, 0.1621, 0.1657, 0.1119, 0.2403, 0.2070, 0.2018,
0.2244, 0.1570, 0.1642, 0.1892, 0.1862, 0.1603, 0.1726,
0.1623, 0.1524, 0.1493, 0.1494, 0.1380, 0.1604, 0.2158,
0.1389, 0.1482, 0.1677, 0.2042, 0.1284, 0.2151, 0.1155,
0.1354, 0.1563], device='cuda:0')), ('layer2.2.bn1.bias', tensor([ 0.1375, 0.0225, -0.0772, 0.0191, 0.0966, 0.0495, -0.0310,
0.0633, -0.0968, 0.0818, -0.0752, -0.0253, -0.0412, 0.0308,
-0.0137, 0.2680, 0.0622, -0.0029, 0.0698, -0.0520, 0.0482,
-0.0226, -0.0084, -0.0744, -0.0868, -0.0652, 0.0399, 0.0512,
0.0236, 0.0933, -0.0298, 0.0088, 0.0235, 0.0286, 0.0584,
0.1792, 0.0735, 0.1293, -0.1163, 0.2136, -0.1089, -0.0553,
0.0383, 0.0160, -0.0720, -0.0421, -0.0587, 0.0376, 0.0621,
-0.0326, 0.0225, -0.0269, 0.0393, -0.0053, -0.0348, 0.0317,
0.0158, 0.0437, -0.0442, -0.0017, 0.0745, 0.0727, -0.0454,
-0.0722, 0.0237, -0.0422, -0.0320, -0.0718, -0.0261, 0.0509,
-0.0768, 0.0078, 0.2535, 0.0508, -0.2020, 0.0159, -0.0641,
-0.0309, -0.0480, -0.0262, 0.0270, 0.0093, -0.0015, -0.0898,
0.0010, 0.0276, 0.0528, -0.0555, 0.0416, -0.0096, -0.0605,
0.0835, 0.0930, 0.0408, 0.0333, -0.0152, -0.0223, 0.0766,
-0.1066, -0.0693, 0.0344, 0.0913, -0.1199, -0.0562, -0.1034,
-0.0769, 0.0707, -0.0081, -0.0790, 0.0056, 0.0025, -0.0215,
-0.0310, 0.0065, -0.0154, -0.0476, 0.1121, 0.0382, -0.0474,
0.0490, -0.0386, 0.0157, -0.1338, 0.0006, -0.1198, 0.0793,
0.0444, 0.0185], device='cuda:0')), ('layer2.2.bn1.running_mean', tensor([-0.0486, -0.1254, -0.0666, 0.0177, -0.2486, -0.0943, -0.1223,
-0.1202, -0.0080, -0.2635, -0.0788, -0.0310, -0.0011, 0.0916,
0.0833, -0.1256, -0.1368, -0.1505, 0.0029, 0.0647, -0.1670,
0.0276, -0.0836, -0.0508, -0.1306, 0.0924, -0.0421, 0.1258,
-0.0017, -0.1482, -0.0619, -0.0592, 0.0222, -0.0273, -0.0895,
-0.0740, 0.0344, 0.1560, 0.0167, -0.1011, 0.0234, -0.1252,
-0.0381, 0.0131, 0.0042, -0.0175, 0.0796, 0.1268, -0.1125,
0.0726, -0.1161, -0.2042, -0.0346, 0.0044, -0.0307, 0.0236,
0.0560, 0.0596, 0.2122, -0.0757, -0.1557, -0.0930, -0.1116,
0.0124, -0.2350, -0.1834, 0.1259, -0.1111, 0.0485, -0.0243,
-0.3510, -0.2041, -0.0341, 0.0025, -0.1354, 0.0210, -0.0669,
-0.0634, -0.0731, 0.0711, 0.1037, 0.1047, -0.0582, -0.0733,
-0.2172, 0.0558, 0.1752, 0.3253, -0.0339, 0.0866, -0.0844,
0.0228, -0.1772, -0.1344, 0.0203, -0.1352, -0.0472, -0.0000,
-0.0320, -0.0742, -0.0655, -0.1024, -0.0183, -0.0718, -0.0814,
-0.0298, -0.0697, -0.0518, 0.0332, -0.1573, -0.0564, -0.0826,
0.1638, 0.1505, -0.0028, -0.0187, 0.0909, -0.0194, -0.0848,
-0.1836, -0.0294, -0.0859, -0.0738, -0.1212, 0.0378, 0.1226,
-0.0117, -0.0120], device='cuda:0')), ('layer2.2.bn1.running_var', tensor([ 0.0268, 0.0235, 0.0196, 0.0378, 0.0223, 0.0279, 0.0219,
0.0211, 0.0333, 0.0459, 0.0169, 0.0199, 0.0194, 0.0385,
0.0259, 0.1163, 0.0378, 0.0101, 0.0258, 0.0191, 0.0328,
0.0236, 0.0309, 0.0265, 0.0114, 0.0227, 0.0328, 0.0259,
0.0274, 0.0430, 0.0318, 0.0366, 0.0185, 0.0277, 0.0408,
0.0538, 0.0179, 0.0192, 0.0203, 0.0762, 0.0159, 0.0440,
0.0234, 0.0232, 0.0263, 0.0233, 0.0234, 0.0235, 0.0324,
0.0183, 0.0223, 0.0148, 0.0213, 0.0235, 0.0366, 0.0326,
0.0306, 0.0277, 0.0208, 0.0276, 0.0216, 0.0488, 0.0247,
0.0135, 0.0515, 0.0264, 0.0283, 0.0181, 0.0217, 0.0265,
0.0317, 0.0382, 0.1040, 0.0335, 0.0109, 0.0166, 0.0173,
0.0301, 0.0218, 0.0168, 0.0405, 0.0274, 0.0382, 0.0305,
0.0303, 0.0309, 0.0193, 0.0216, 0.0140, 0.0299, 0.0246,
0.0421, 0.0163, 0.0234, 0.0280, 0.0213, 0.0296, 0.0279,
0.0190, 0.0115, 0.0431, 0.0185, 0.0286, 0.0375, 0.0236,
0.0325, 0.0324, 0.0205, 0.0188, 0.0380, 0.0256, 0.0167,
0.0218, 0.0174, 0.0209, 0.0237, 0.0437, 0.0234, 0.0297,
0.0346, 0.0189, 0.0246, 0.0134, 0.0256, 0.0172, 0.0192,
0.0160, 0.0228], device='cuda:0')), ('layer2.2.conv2.weight', tensor([[[[-3.1302e-03, -3.1556e-03, 3.8333e-04],
[ 1.0679e-02, -3.3345e-03, 1.0498e-02],
[ 1.7142e-03, -8.1773e-03, 4.2616e-03]],
[[ 1.6654e-02, 9.1589e-03, 1.0449e-02],
[ 1.4159e-02, 1.4123e-02, -5.6379e-03],
[ 1.0631e-02, 1.1550e-02, 1.4759e-02]],
[[ 1.3936e-02, 1.1762e-02, 7.6583e-03],
[ 5.1115e-03, -2.7663e-03, 2.6938e-02],
[-7.7951e-04, 6.5254e-03, -5.8126e-03]],
...,
[[ 9.4068e-03, 3.3847e-04, -6.2770e-03],
[-1.9364e-03, -5.7068e-03, -5.2139e-03],
[-1.5536e-02, 3.0893e-03, 8.7088e-04]],
[[-2.6588e-02, -1.3738e-02, 1.9362e-03],
[-1.1497e-02, 5.4170e-03, 2.2290e-02],
[-1.5209e-02, 1.0042e-02, 1.4244e-02]],
[[ 2.0787e-02, 1.2619e-02, 8.0365e-03],
[ 2.5076e-02, 1.5249e-02, 3.9060e-02],
[ 1.4655e-02, 1.3425e-02, 2.6900e-02]]],
[[[-1.3546e-02, 1.0102e-02, -5.6306e-03],
[ 1.8917e-03, 4.0935e-02, 2.9032e-04],
[-3.2169e-02, -5.6877e-02, -3.7226e-02]],
[[-9.4402e-03, -1.2173e-02, 2.2989e-03],
[ 5.8581e-04, -1.7501e-03, 1.4888e-02],
[ 4.2108e-03, -8.8788e-04, 1.6264e-02]],
[[ 9.6043e-03, 3.3679e-03, -9.1322e-03],
[-3.7454e-03, -2.5192e-02, -3.1959e-02],
[-1.4426e-02, -4.6583e-03, -1.8655e-02]],
...,
[[-1.9185e-02, -1.0943e-02, -1.8933e-02],
[-1.2623e-02, 3.7866e-03, -1.0680e-02],
[ 2.0189e-03, 7.9173e-03, 2.6661e-02]],
[[ 1.3142e-02, -2.9681e-03, -6.9075e-03],
[ 2.0208e-02, 2.5027e-03, -1.6389e-02],
[ 1.4347e-03, -7.2946e-03, -1.0424e-02]],
[[-1.7231e-02, -6.5504e-03, -1.3695e-02],
[-1.1624e-02, 4.5325e-03, 1.2338e-03],
[-7.9682e-03, 4.0773e-04, -1.0477e-03]]],
[[[-1.4542e-02, -2.2976e-02, -1.3867e-02],
[-6.8874e-05, 6.3162e-03, 7.4062e-03],
[-5.3631e-03, -1.9375e-03, 8.8525e-03]],
[[-7.8354e-03, -9.8175e-03, -9.5295e-03],
[-9.0836e-03, -7.8326e-03, 3.8246e-03],
[ 1.2576e-02, 2.6117e-02, -1.9496e-03]],
[[ 3.4312e-03, 1.4886e-02, 1.9845e-02],
[-1.7894e-02, -1.6905e-02, -5.7597e-03],
[ 6.9308e-03, -2.9813e-03, -1.3229e-02]],
...,
[[-6.5217e-03, -1.5325e-02, 1.5252e-02],
[ 2.3357e-02, -4.1022e-02, -4.0224e-03],
[ 1.3244e-02, 4.9131e-02, -1.4938e-02]],
[[-2.7648e-03, -1.0411e-02, 5.6259e-04],
[-1.8668e-03, 3.0416e-02, 7.0544e-03],
[-2.5378e-02, -1.2816e-02, 1.4620e-02]],
[[-2.3961e-02, -4.3054e-02, -1.8946e-02],
[ 4.6608e-02, 8.2388e-03, 2.4859e-02],
[ 1.4446e-02, 1.1414e-02, 4.3087e-02]]],
...,
[[[-1.2832e-02, -9.8999e-03, -9.7872e-03],
[-1.4953e-02, -3.4269e-03, -1.3721e-02],
[-2.8918e-02, 5.7522e-02, 1.9740e-02]],
[[-8.8964e-03, -1.9896e-02, -1.0263e-02],
[-9.1614e-03, -1.4733e-02, -2.9040e-03],
[-3.1857e-02, 4.7644e-02, 5.3215e-02]],
[[-2.4541e-02, -1.3963e-02, -1.5121e-02],
[-2.0705e-02, -1.2160e-02, -2.5758e-02],
[ 1.2093e-02, 9.5702e-03, 1.2815e-02]],
...,
[[ 1.4657e-02, 2.1536e-02, 1.2741e-03],
[ 6.3637e-03, 5.7634e-04, 1.0379e-02],
[ 7.6955e-03, -1.1966e-02, 6.8439e-04]],
[[ 4.8690e-03, 4.9937e-03, 1.3563e-03],
[-1.6994e-02, -2.6711e-02, -2.6102e-02],
[-1.1411e-02, -2.1785e-02, -1.4453e-02]],
[[ 2.4642e-02, 1.8539e-02, -3.5493e-04],
[ 2.0089e-02, 1.7848e-02, -7.1624e-03],
[ 1.9272e-02, -1.4525e-02, -4.6722e-02]]],
[[[ 2.2050e-02, 1.9969e-02, 1.4083e-02],
[ 1.4105e-02, -1.0612e-02, -8.6998e-03],
[ 1.2431e-03, 1.5315e-02, -6.0682e-03]],
[[ 2.2922e-04, -6.6886e-03, 1.1839e-03],
[ 8.1434e-04, 4.4861e-03, -6.8180e-03],
[-2.7884e-03, 3.1447e-02, 2.0861e-02]],
[[ 1.0742e-02, -2.3188e-02, -8.3067e-03],
[ 8.8557e-02, 3.8402e-03, -1.1184e-01],
[ 2.5634e-02, -2.0695e-02, -2.0699e-02]],
...,
[[ 3.1371e-03, -1.8001e-02, -1.7856e-02],
[-1.1404e-02, 9.1344e-03, 3.5242e-02],
[ 8.3053e-04, 1.2192e-02, 8.2051e-03]],
[[ 2.5233e-02, 9.5351e-04, -6.2553e-03],
[-1.5711e-02, -2.0540e-02, -3.1693e-02],
[-6.6347e-03, -1.9945e-02, -1.5864e-02]],
[[-9.2498e-03, 5.4429e-03, 3.4973e-03],
[-4.1350e-02, 1.3857e-02, 2.2276e-02],
[ 7.5377e-03, 1.8090e-02, 2.7004e-02]]],
[[[-3.6390e-03, -1.6915e-02, -1.4160e-02],
[ 3.0303e-03, 1.6015e-03, 9.6799e-03],
[ 1.8470e-02, 2.3476e-02, 1.8043e-02]],
[[-1.8460e-02, -4.0362e-02, -1.1782e-02],
[-1.6825e-02, 1.9933e-02, 1.5326e-03],
[-6.4067e-04, -1.1057e-02, 5.4452e-03]],
[[ 4.8412e-03, 1.7270e-03, 8.2047e-03],
[ 6.2839e-03, -1.5603e-02, -1.3443e-02],
[ 1.8032e-02, 3.6521e-02, 3.5794e-02]],
...,
[[-4.0975e-02, -6.5356e-02, -1.6572e-02],
[-7.1748e-02, 2.4892e-02, -1.2404e-02],
[-5.6559e-03, -3.4650e-03, 2.3677e-02]],
[[ 2.8632e-03, 2.8907e-02, 4.2334e-03],
[ 1.6220e-02, -1.9336e-02, 3.0459e-04],
[-2.2507e-02, 1.5587e-02, 9.9927e-03]],
[[-9.9456e-03, 9.8812e-03, 2.4514e-03],
[-2.0043e-02, -2.1091e-03, -2.1618e-02],
[ 1.1337e-02, 1.8307e-02, 1.1379e-02]]]], device='cuda:0')), ('layer2.2.bn2.weight', tensor([ 0.2214, 0.1721, 0.2176, 0.1959, 0.1742, 0.2177, 0.1805,
0.2151, 0.1563, 0.1343, 0.1279, 0.1864, 0.1404, 0.1347,
0.2264, 0.2030, 0.2185, 0.1454, 0.1638, 0.2064, 0.1385,
0.1764, 0.1671, 0.1952, 0.2158, 0.1879, 0.1612, 0.1774,
0.1355, 0.2028, 0.1216, 0.2157, 0.2119, 0.1582, 0.1622,
0.2311, 0.1990, 0.1647, 0.1607, 0.1608, 0.2214, 0.1909,
0.1611, 0.2305, 0.1727, 0.1899, 0.1830, 0.2176, 0.1987,
0.1904, 0.1906, 0.2067, 0.2098, 0.1184, 0.2161, 0.1449,
0.2168, 0.1576, 0.1104, 0.2230, 0.2038, 0.2066, 0.1992,
0.1962, 0.1973, 0.1898, 0.1507, 0.1287, 0.1824, 0.2102,
0.1911, 0.1703, 0.1391, 0.1661, 0.1218, 0.2001, 0.2231,
0.1790, 0.1083, 0.1760, 0.2181, 0.1710, 0.1995, 0.1270,
0.1417, 0.1999, 0.1815, 0.1847, 0.1651, 0.2537, 0.2222,
0.2202, 0.1587, 0.2057, 0.1708, 0.1237, 0.2065, 0.2275,
0.1513, 0.1368, 0.2161, 0.2136, 0.1753, 0.2129, 0.1749,
0.1589, 0.1498, 0.1623, 0.1935, 0.2039, 0.1690, 0.2222,
0.2065, 0.1721, 0.1346, 0.2136, 0.1893, 0.2136, 0.2059,
0.1386, 0.2209, 0.1666, 0.1613, 0.2016, 0.1986, 0.2145,
0.2138, 0.1231], device='cuda:0')), ('layer2.2.bn2.bias', tensor([-0.0689, -0.0313, -0.0733, -0.0350, -0.0190, -0.0620, 0.0035,
0.0132, -0.0292, 0.2037, 0.1439, -0.0762, 0.0221, 0.1385,
-0.0421, -0.0142, -0.0809, 0.0121, -0.0230, -0.0762, 0.0541,
-0.0423, -0.0161, -0.0310, -0.0601, 0.0999, 0.0646, 0.1196,
0.0753, -0.1081, 0.1915, -0.0441, -0.0390, 0.0091, -0.0384,
-0.0733, -0.0219, 0.0743, -0.0316, 0.0027, -0.0706, -0.0089,
0.0214, -0.0509, 0.0323, -0.0528, -0.0301, -0.1541, -0.0387,
-0.0589, -0.0495, -0.0700, -0.0180, 0.0271, -0.1037, -0.0268,
-0.1122, 0.0781, 0.2120, -0.0242, -0.0506, -0.0394, -0.0469,
-0.1692, -0.0786, 0.0449, 0.0841, 0.2805, -0.0102, -0.0755,
0.0087, -0.0163, 0.0084, -0.0440, 0.1122, -0.0906, -0.0639,
0.0545, 0.1696, 0.0667, -0.0352, -0.0517, -0.0681, 0.1267,
0.1488, -0.0290, -0.0383, -0.0328, 0.0788, -0.1638, -0.1247,
-0.0539, -0.0066, 0.0618, 0.0331, 0.0680, -0.0730, -0.1106,
0.0354, 0.0932, -0.0082, -0.1190, -0.0311, -0.0941, -0.0076,
0.1077, 0.0925, 0.0727, 0.0321, 0.0552, -0.0851, -0.0713,
-0.0749, 0.0496, 0.0787, -0.1726, 0.0821, -0.1025, -0.0183,
0.0164, -0.1447, 0.0868, 0.0499, -0.0962, -0.0720, -0.0975,
-0.0382, 0.1936], device='cuda:0')), ('layer2.2.bn2.running_mean', tensor([-0.0598, 0.0115, -0.0593, 0.0467, -0.0997, -0.1773, 0.1655,
0.0408, -0.0094, 0.0201, 0.0931, -0.2313, -0.0581, 0.0878,
0.1587, 0.0479, -0.0768, 0.0142, -0.2025, -0.1242, -0.0824,
-0.0712, -0.0706, 0.1304, -0.0839, 0.1789, 0.1900, 0.2495,
-0.0501, -0.1121, -0.3133, 0.1271, -0.1171, -0.0251, -0.1486,
-0.0764, 0.0844, -0.0658, -0.0455, 0.0037, 0.1186, 0.0706,
-0.0856, -0.0109, 0.0989, 0.0485, -0.0461, -0.1100, 0.0636,
-0.0875, -0.0390, 0.0603, 0.0412, -0.1436, -0.1237, 0.0690,
-0.0766, 0.0447, 0.0737, 0.0542, -0.1584, -0.1193, 0.0086,
0.0182, -0.0633, 0.1961, 0.0564, -0.1393, 0.1121, -0.0754,
-0.0932, -0.0737, -0.0764, 0.1731, -0.1006, -0.1147, -0.1893,
0.0358, -0.1083, 0.1174, 0.0714, -0.0350, -0.0600, 0.1615,
0.0672, -0.0751, -0.1035, -0.0733, 0.2198, 0.0070, -0.0951,
-0.0762, 0.0240, 0.1475, 0.1515, -0.1642, -0.1112, 0.0069,
-0.3446, -0.1314, 0.0220, -0.0039, 0.1072, -0.0520, -0.0344,
0.1267, 0.1098, 0.2006, 0.1288, 0.1078, -0.1614, -0.0472,
-0.1547, -0.1920, -0.1662, 0.1391, 0.1256, -0.0007, 0.0817,
0.0260, -0.1715, 0.1747, -0.1896, -0.1408, 0.0148, -0.0044,
0.0791, -0.1543], device='cuda:0')), ('layer2.2.bn2.running_var', tensor(1.00000e-02 *
[ 2.3122, 1.3506, 2.1545, 1.8420, 2.4524, 3.3761, 2.7163,
2.8899, 1.5107, 2.6212, 1.6568, 1.2806, 1.0481, 2.0796,
3.6430, 2.6995, 2.1856, 0.9302, 1.1140, 1.4490, 1.2032,
1.5976, 1.1571, 1.9631, 2.1637, 4.8277, 2.4767, 4.5362,
2.2598, 3.0386, 1.6297, 3.4322, 2.3071, 1.4067, 1.2297,
2.4349, 2.0147, 1.8916, 1.5594, 2.0022, 2.8407, 2.0501,
2.0938, 3.0104, 2.8363, 1.7389, 1.3707, 1.4718, 2.3899,
2.0484, 1.9319, 1.7144, 2.6828, 0.8792, 1.4183, 1.3901,
3.2452, 3.1968, 1.3576, 3.4623, 2.6049, 3.2984, 1.6533,
1.1792, 1.3561, 3.3851, 2.1610, 2.2667, 2.0088, 1.9632,
2.7376, 1.9447, 3.0756, 1.2838, 1.5414, 1.9388, 3.8768,
2.7923, 1.1360, 2.3843, 2.2282, 1.3047, 1.6421, 2.2289,
2.3117, 2.0094, 1.5850, 1.4084, 2.8760, 1.0326, 1.6725,
2.6477, 1.7103, 3.9172, 2.9911, 1.7148, 1.9727, 1.9594,
2.3749, 1.6972, 2.9680, 2.1783, 1.3108, 1.9240, 1.9365,
2.5777, 1.9505, 2.9460, 2.2705, 5.5931, 0.7255, 2.9636,
2.3596, 2.3255, 1.9298, 1.0020, 4.5591, 1.7011, 2.0744,
1.3651, 1.1797, 2.9305, 1.7633, 1.3466, 1.1975, 1.8808,
2.6639, 2.0136], device='cuda:0')), ('layer2.2.conv3.weight', tensor([[[[-1.7472e-02]],
[[ 2.6974e-02]],
[[-6.1007e-03]],
...,
[[-4.4693e-03]],
[[-2.8153e-04]],
[[ 1.4522e-02]]],
[[[ 2.0458e-03]],
[[-1.0016e-02]],
[[-3.2662e-03]],
...,
[[ 2.6745e-02]],
[[-1.7600e-02]],
[[-2.3394e-02]]],
[[[-2.1447e-02]],
[[ 3.7622e-03]],
[[ 1.4855e-02]],
...,
[[-1.2352e-02]],
[[-1.1636e-04]],
[[-5.3608e-04]]],
...,
[[[ 2.4283e-02]],
[[-2.2397e-02]],
[[-1.1301e-02]],
...,
[[ 6.5993e-03]],
[[ 6.2060e-03]],
[[ 1.1466e-02]]],
[[[ 5.3356e-03]],
[[ 7.6939e-03]],
[[ 1.3788e-02]],
...,
[[-5.2157e-03]],
[[ 8.2105e-03]],
[[ 6.3159e-03]]],
[[[ 7.1847e-03]],
[[ 3.5737e-03]],
[[ 1.9639e-02]],
...,
[[-1.3127e-02]],
[[-6.6329e-03]],
[[ 1.2715e-03]]]], device='cuda:0')), ('layer2.2.bn3.weight', tensor([ 0.0572, 0.0513, 0.0701, 0.2007, 0.2130, 0.1163, 0.0012,
0.1823, 0.1187, 0.1230, 0.1162, 0.0789, 0.0843, 0.0609,
0.1762, 0.0784, 0.0554, 0.1829, 0.0934, 0.1814, 0.0428,
0.0973, 0.0335, 0.1376, 0.0642, 0.2001, 0.1781, 0.1554,
0.1482, 0.1378, 0.0466, 0.1781, 0.0446, 0.1049, 0.0964,
0.0300, 0.0043, 0.0754, 0.2078, 0.0043, 0.0759, 0.0932,
0.2199, 0.0826, 0.1367, 0.1090, 0.0942, 0.1007, 0.0665,
0.0806, 0.1017, 0.1144, 0.2540, 0.0861, 0.0667, 0.0575,
0.1358, 0.0983, 0.0041, 0.2154, 0.0973, 0.0605, 0.1822,
0.0933, 0.0918, 0.1295, 0.1022, 0.0661, 0.0623, 0.0581,
0.0637, 0.1190, 0.0459, 0.1767, 0.0168, 0.1007, 0.0239,
0.0068, 0.0625, 0.0865, 0.0413, 0.0604, 0.0558, 0.1156,
0.1678, 0.0521, 0.0745, 0.1013, 0.0580, 0.2115, 0.0911,
0.0517, 0.0509, 0.1069, 0.1632, 0.1126, 0.0647, 0.0572,
0.0580, 0.0552, 0.1158, 0.0740, -0.0273, -0.0082, 0.2625,
0.1106, 0.1173, 0.1051, 0.0448, 0.1251, 0.0505, 0.1710,
0.2576, 0.1376, 0.0867, 0.0014, 0.0809, 0.0327, 0.0336,
0.1216, 0.1788, 0.1153, 0.0822, 0.1186, 0.2659, 0.1741,
0.0052, 0.0337, 0.1619, 0.1166, 0.1025, 0.1151, 0.1388,
0.0770, 0.0053, 0.2740, 0.0846, 0.1296, 0.1493, 0.1267,
0.0597, 0.1591, 0.1063, 0.1961, 0.0483, 0.0920, 0.1002,
0.0645, 0.0543, 0.1088, 0.0278, 0.1533, 0.1285, 0.1820,
0.1611, 0.0110, 0.1421, 0.0983, 0.0916, 0.1161, 0.1526,
0.0565, 0.0151, 0.0767, 0.0427, 0.2274, 0.0219, 0.2281,
0.0649, 0.1308, 0.1560, -0.0069, 0.0039, -0.0311, 0.1364,
0.0609, 0.0975, 0.0395, 0.1814, 0.0876, 0.1118, 0.1000,
0.0369, 0.0798, 0.0906, 0.0694, 0.1220, 0.0516, 0.0508,
0.1456, 0.0288, 0.0925, 0.1705, 0.1075, 0.1666, 0.0618,
0.0837, 0.0783, 0.1867, 0.0379, 0.3287, 0.1095, 0.0732,
0.2327, 0.1283, 0.1315, 0.1142, 0.0504, 0.0154, 0.3249,
0.0779, 0.0358, 0.0198, 0.1475, 0.1205, 0.0879, 0.2058,
0.0115, 0.0814, 0.0804, 0.0946, 0.1902, 0.0837, 0.1229,
0.0143, 0.1563, 0.0703, 0.1395, 0.0855, 0.0359, 0.1170,
0.1338, 0.1362, 0.2025, 0.1494, 0.0991, 0.0771, 0.0020,
-0.0063, 0.1195, 0.0837, 0.1724, 0.0672, 0.0377, 0.1593,
0.0878, 0.0502, 0.0175, 0.0965, 0.1385, 0.1657, 0.1162,
0.1028, 0.0725, 0.1250, 0.1454, -0.0084, 0.0637, 0.1134,
0.0136, 0.1180, 0.0247, 0.0537, -0.0115, 0.1539, -0.0120,
0.3069, 0.0624, 0.0089, 0.0784, 0.1168, 0.1698, 0.1871,
0.0929, 0.1473, 0.0847, 0.1326, 0.1473, 0.1228, 0.1336,
0.0743, 0.2222, 0.0883, 0.1828, 0.1132, 0.0704, 0.1213,
0.0902, 0.0415, 0.1138, 0.0352, 0.2897, 0.0916, 0.0631,
0.0687, 0.0172, 0.1139, 0.1199, 0.1306, 0.1355, 0.1588,
0.0467, 0.1177, 0.1674, 0.0644, 0.0893, 0.2456, 0.1993,
0.1414, 0.0956, 0.0945, 0.0540, 0.1131, 0.1612, 0.1691,
0.0390, 0.1526, 0.1769, 0.1023, 0.1605, 0.0868, 0.0815,
0.1840, 0.2672, 0.0122, 0.1325, 0.1417, 0.1323, 0.1021,
0.1256, 0.0109, 0.0460, 0.1077, 0.1256, 0.1203, 0.1260,
0.0761, 0.0358, 0.0550, 0.1311, 0.1372, 0.1867, 0.1151,
0.1249, 0.1045, 0.0580, 0.1403, 0.0596, 0.1390, -0.0181,
0.0823, 0.1477, 0.0617, 0.0796, 0.1584, 0.0956, 0.1101,
0.1017, 0.0879, 0.1078, 0.0771, 0.0438, 0.1673, 0.0882,
0.1667, 0.0447, 0.1636, -0.0518, 0.0370, 0.0470, 0.0539,
0.0540, 0.0883, 0.0215, 0.0168, 0.0360, 0.1388, 0.1197,
0.2181, 0.0667, -0.0053, 0.0495, 0.0450, 0.1803, 0.0378,
0.1474, 0.1313, 0.0679, 0.0353, 0.0910, 0.0611, 0.0460,
0.0867, 0.0864, 0.0310, 0.1489, 0.0776, 0.1239, 0.0854,
0.1083, 0.1216, 0.0186, 0.0180, 0.1009, 0.1910, 0.0753,
0.0302, 0.1190, 0.1746, 0.2461, 0.0172, 0.0771, 0.1705,
0.2232, 0.0322, 0.0642, 0.2283, 0.0553, 0.1161, 0.1704,
0.1194, 0.0243, 0.0603, 0.0632, 0.1568, 0.2214, 0.0624,
0.0438, 0.0822, 0.1046, 0.0695, 0.1556, 0.0564, 0.0446,
0.1318, 0.0841, 0.1011, 0.0388, 0.1426, 0.1567, 0.1475,
0.0517, 0.0927, 0.1638, 0.1908, 0.0565, 0.1558, 0.0968,
0.0198, 0.1799, 0.1659, 0.2310, 0.1778, 0.0523, 0.1440,
0.1129, 0.0765, 0.0757, 0.0904, 0.0721, 0.0746, 0.0185,
0.1771, 0.1030, 0.2082, 0.0926, 0.0816, 0.0759, 0.0496,
0.0541, 0.1026, 0.0844, 0.1304, 0.0859, 0.1788, 0.0786,
-0.0647, 0.1403, 0.0357, 0.0527, 0.1047, 0.1177, 0.1492,
0.0794, 0.0375, 0.1701, 0.0531, 0.0670, 0.1271, 0.0515,
0.0727, 0.1686, 0.1360, 0.0909, 0.0615, 0.1081, 0.0807,
0.1218, 0.1666, 0.0098, 0.1504, 0.0429, 0.1308, 0.1624,
0.1157, 0.0948, 0.0596, 0.2338, 0.0812, 0.0434, 0.0457,
0.0894], device='cuda:0')), ('layer2.2.bn3.bias', tensor([-0.0656, 0.0623, -0.1095, -0.0543, -0.1288, 0.0748, -0.0087,
-0.2070, -0.1142, -0.0899, -0.0936, -0.0628, -0.1292, 0.0590,
-0.0655, -0.0571, 0.0550, -0.0526, -0.0630, 0.0080, -0.0005,
-0.0885, -0.0307, -0.1267, -0.0520, -0.0959, -0.0952, 0.0678,
-0.0959, -0.0740, -0.0278, -0.1276, -0.0307, -0.1096, -0.0940,
0.0040, -0.0002, -0.0428, 0.0963, 0.0013, -0.0750, -0.0390,
-0.0059, -0.1175, -0.0447, -0.0469, -0.1235, -0.0833, -0.0262,
0.0108, -0.1414, 0.0294, 0.0362, -0.0755, -0.1037, 0.0097,
0.0370, 0.0668, -0.0107, -0.0486, -0.1051, -0.0508, -0.0455,
-0.1553, -0.0128, -0.0391, -0.0374, -0.0178, 0.0127, -0.0422,
-0.0403, -0.0207, -0.0456, -0.0370, 0.0010, -0.1166, -0.0340,
-0.1577, -0.1075, -0.0915, -0.0241, -0.0279, -0.0631, -0.0276,
-0.0567, -0.0367, -0.0473, -0.0849, -0.0920, -0.0569, -0.0460,
-0.0465, -0.0230, -0.1320, -0.1337, -0.0459, -0.0368, -0.0858,
-0.0256, 0.0244, -0.1281, -0.0437, -0.0698, 0.0004, -0.2142,
0.0014, -0.0459, -0.0186, -0.0032, -0.0119, -0.0258, -0.0281,
0.0127, -0.0855, -0.0993, -0.0281, -0.0504, -0.0157, -0.0182,
-0.0746, -0.0540, -0.0915, -0.0878, -0.0552, -0.0401, -0.2473,
-0.0075, 0.0006, -0.1237, 0.0053, -0.0039, -0.0345, -0.0493,
-0.0154, 0.0169, -0.0434, -0.0039, -0.1301, -0.0935, -0.0836,
-0.0659, 0.0098, -0.1002, -0.0746, 0.0687, -0.1072, -0.0738,
-0.0420, -0.0593, -0.0801, -0.0153, -0.1203, -0.0517, -0.0463,
-0.1568, -0.0071, -0.1911, -0.0159, -0.1125, -0.0606, -0.0200,
-0.0475, -0.0016, -0.1033, -0.0176, -0.0340, -0.0299, -0.0067,
-0.0590, -0.0333, -0.1070, 0.0107, 0.0087, 0.0058, 0.0171,
-0.0699, -0.1343, -0.0112, -0.0922, -0.0580, -0.1174, -0.0538,
-0.0270, -0.1041, -0.0816, -0.1074, -0.1203, -0.0518, -0.1062,
0.1657, -0.0258, -0.0724, -0.1844, -0.1733, -0.0928, -0.0231,
-0.0777, -0.0377, -0.0879, -0.0200, 0.0291, -0.1082, -0.0710,
-0.1176, -0.1808, -0.0343, -0.0676, -0.0296, 0.0085, 0.0265,
-0.0141, -0.0271, -0.0298, -0.1274, 0.0342, -0.0807, -0.0610,
-0.0050, -0.1240, -0.0361, -0.0976, -0.0336, -0.0889, -0.0407,
-0.0147, -0.0681, -0.0429, -0.1204, -0.0312, -0.0209, -0.0666,
-0.1509, -0.1203, -0.0113, -0.1406, 0.0105, -0.0592, -0.0009,
-0.0041, -0.0572, -0.0998, -0.2547, -0.0238, -0.0774, -0.0390,
-0.0858, -0.0564, -0.0048, -0.0580, -0.0574, -0.0793, -0.1391,
-0.0698, -0.0473, -0.0204, -0.1036, -0.0036, -0.0555, 0.0171,
-0.0010, -0.0866, -0.0006, -0.0281, -0.0358, -0.0767, -0.0007,
0.0130, -0.1123, -0.0186, -0.0881, 0.0657, -0.0644, -0.0657,
-0.0724, -0.0484, -0.0282, -0.0598, -0.0328, -0.0359, -0.0460,
-0.0572, -0.0811, 0.0089, -0.0568, -0.0813, -0.0852, -0.0294,
-0.0569, -0.0049, -0.0644, 0.0474, -0.0951, -0.0773, -0.0731,
-0.0529, -0.0084, -0.0034, -0.1180, -0.1336, 0.0490, -0.1214,
-0.0563, -0.0184, -0.0599, -0.0625, -0.0171, -0.0155, -0.0850,
0.0478, -0.1303, -0.1027, -0.0287, -0.1192, -0.1356, -0.1505,
-0.0838, 0.0006, -0.2332, -0.1373, -0.0274, -0.0007, -0.0795,
0.1128, -0.0532, -0.0032, -0.2691, -0.1173, -0.0292, -0.0591,
-0.1105, -0.0044, -0.0189, -0.1673, -0.1041, -0.0846, -0.0492,
-0.0390, -0.0709, 0.0135, 0.0003, -0.1606, 0.0275, -0.1237,
-0.0938, -0.1119, -0.0370, -0.1266, -0.0587, 0.0407, -0.0700,
-0.0517, 0.0351, -0.0505, -0.0921, -0.2230, -0.0779, -0.0945,
-0.1070, -0.0558, -0.0325, -0.0540, -0.0275, -0.0892, -0.1113,
0.0114, -0.0397, 0.0964, -0.0294, -0.0294, -0.0181, -0.0070,
-0.0693, -0.0592, -0.0170, -0.0032, -0.0309, -0.0706, -0.0504,
0.0011, -0.0789, 0.0048, -0.0064, -0.0959, -0.1322, -0.0162,
-0.0680, -0.0693, -0.1036, 0.0471, -0.0888, -0.0303, -0.0452,
-0.1000, -0.0800, -0.0333, -0.0536, -0.1074, -0.0422, -0.0190,
-0.0688, -0.1408, -0.0471, -0.0313, -0.0708, -0.2162, -0.0389,
-0.0847, 0.0167, -0.0667, -0.0491, -0.0157, -0.0647, -0.1543,
-0.0526, -0.0474, -0.0306, -0.0212, -0.0108, -0.1527, -0.0932,
-0.0771, -0.0352, -0.0696, -0.0689, -0.1471, -0.0780, -0.0958,
-0.0506, -0.0843, -0.0394, -0.1198, 0.0236, -0.0374, -0.0605,
-0.1516, -0.0415, -0.1013, 0.0375, -0.0141, -0.0659, 0.0275,
-0.0584, -0.1379, -0.0271, 0.0199, 0.0512, -0.1530, -0.0874,
0.0169, -0.0876, -0.1972, 0.0223, -0.0600, 0.0066, -0.0800,
-0.0635, -0.0453, -0.0803, -0.0316, -0.1109, -0.0635, -0.0078,
0.0633, -0.0805, -0.2414, -0.0595, 0.0132, -0.0745, -0.0975,
-0.0193, -0.0836, -0.1500, -0.1593, -0.0926, -0.1080, -0.0348,
-0.0726, -0.0226, -0.0030, -0.0651, 0.0457, -0.0648, -0.0374,
-0.0789, -0.0342, -0.1050, -0.0697, -0.0372, -0.1122, -0.1033,
0.0137, 0.1496, -0.0535, -0.0457, -0.1078, -0.0657, -0.0461,
-0.0396, -0.1002, 0.0032, -0.0886, -0.0430, -0.1154, -0.0426,
-0.0749, -0.1382, -0.0942, -0.0085, 0.0473, -0.0135, -0.0202,
-0.1279], device='cuda:0')), ('layer2.2.bn3.running_mean', tensor([-0.0089, -0.0187, 0.0055, -0.0432, -0.0250, 0.0370, 0.0003,
0.0022, 0.0035, 0.0016, -0.0256, -0.0602, -0.0340, 0.0081,
-0.0527, -0.0223, 0.0404, -0.0080, 0.0207, -0.0236, 0.0099,
0.0166, 0.0098, 0.0121, -0.0311, 0.0030, 0.0321, -0.0057,
0.0112, -0.0431, -0.0124, 0.0497, 0.0020, 0.0515, 0.0090,
-0.0133, -0.0081, -0.0129, 0.0315, -0.0225, 0.0087, -0.0176,
-0.0334, -0.0009, -0.0045, -0.0294, -0.0031, -0.0207, 0.0046,
-0.0279, -0.0029, -0.0248, 0.0429, -0.0221, 0.0099, 0.0196,
-0.0570, -0.0341, -0.0018, -0.0097, -0.0111, -0.0041, -0.0329,
-0.0001, 0.0006, -0.0296, 0.0342, -0.0168, 0.0157, -0.0138,
0.0292, -0.0420, 0.0279, 0.0066, 0.0061, 0.0489, 0.0019,
0.0012, 0.0059, 0.0201, 0.0255, 0.0182, 0.0077, -0.0004,
-0.0262, -0.0057, 0.0232, 0.0452, -0.0129, 0.0099, -0.0133,
-0.0019, 0.0019, 0.0477, 0.0587, -0.0183, 0.0035, 0.0181,
-0.0240, -0.0098, 0.0180, -0.0221, -0.0034, 0.0054, -0.0249,
-0.0205, 0.0130, -0.0398, -0.0166, 0.0014, 0.0327, -0.0087,
-0.0372, -0.0509, 0.0298, 0.0013, 0.0610, -0.0109, 0.0182,
-0.0457, -0.0147, -0.0120, -0.0023, 0.0174, -0.0043, 0.0334,
0.0022, 0.0324, 0.0051, 0.0358, -0.0384, 0.0217, -0.0550,
0.0066, -0.0194, -0.0074, -0.0154, 0.0909, 0.0036, -0.0200,
-0.0094, 0.0091, -0.0033, 0.0067, -0.0013, 0.0311, -0.0061,
-0.0446, 0.0433, -0.0146, 0.0142, 0.0167, 0.0296, -0.0054,
0.0290, -0.0150, 0.0338, 0.0567, 0.0313, 0.0086, -0.0156,
0.0078, -0.0057, 0.0356, 0.0349, 0.0396, -0.0319, -0.0012,
0.0232, 0.0182, 0.0240, -0.0029, 0.0025, -0.0109, 0.0619,
0.0025, -0.0388, -0.0024, -0.0282, 0.0131, 0.0217, 0.0060,
0.0180, 0.0298, 0.0034, 0.0243, 0.0075, 0.0035, 0.0402,
0.0421, 0.0056, -0.0166, -0.0275, 0.0619, 0.0160, -0.0477,
-0.0121, 0.0035, -0.0416, 0.0088, -0.0172, -0.0171, 0.0171,
0.0437, 0.0129, -0.0296, -0.0366, -0.0234, 0.0039, -0.0087,
-0.0285, 0.0228, 0.0205, -0.0004, 0.0044, -0.0561, -0.0171,
0.0056, 0.0355, -0.0114, -0.0075, 0.0133, -0.0090, -0.0151,
0.0162, 0.0098, -0.0031, 0.0084, -0.0080, 0.0254, 0.0147,
-0.0135, -0.0170, -0.0098, 0.0064, -0.0421, 0.0082, 0.0128,
-0.0060, 0.0037, 0.0135, 0.0619, 0.0198, 0.0078, 0.0132,
-0.0660, 0.0529, 0.0083, -0.0329, -0.0843, -0.0230, 0.0450,
-0.0220, -0.0081, 0.0022, 0.0106, -0.0029, -0.0201, -0.0410,
-0.0175, -0.0610, -0.0150, 0.0031, -0.0176, 0.0014, -0.0063,
0.0366, 0.0244, -0.0119, -0.0325, 0.0201, -0.0394, 0.0128,
0.0034, -0.0281, -0.0040, 0.0083, 0.0432, -0.0361, -0.0263,
-0.0034, 0.0268, -0.0038, -0.0063, -0.0257, 0.0348, -0.0268,
-0.0143, 0.0297, -0.0232, -0.0067, 0.0437, -0.0276, 0.0195,
0.0012, -0.0189, -0.0243, -0.0024, 0.0254, -0.0091, 0.0062,
-0.0242, -0.0196, -0.0404, -0.0456, -0.0184, -0.0227, 0.0742,
0.0202, -0.0067, -0.0150, 0.0399, -0.0192, 0.0308, -0.0115,
0.0096, 0.0008, 0.0097, 0.0117, -0.0189, -0.0396, -0.0081,
0.0390, 0.0030, -0.0060, 0.0175, -0.0159, -0.0270, -0.0224,
-0.0185, 0.0059, 0.0348, 0.0083, -0.0142, 0.0344, -0.0019,
-0.0145, 0.0003, -0.0119, 0.0076, 0.0118, -0.0235, -0.0050,
-0.0290, -0.0150, -0.0164, 0.0184, -0.0071, 0.0081, 0.0684,
-0.0132, 0.0896, -0.0107, -0.0220, 0.0389, -0.0084, 0.0114,
0.0255, -0.0107, -0.0335, -0.0033, -0.0225, -0.0257, 0.0139,
-0.0385, 0.0140, 0.0321, 0.0145, 0.0268, 0.0175, 0.0321,
0.0282, -0.0230, 0.0009, 0.0006, 0.0375, -0.0121, 0.0293,
0.0058, -0.0286, 0.0006, -0.0351, -0.0026, 0.0373, -0.0195,
-0.0272, -0.0025, -0.0357, 0.0039, -0.0002, -0.0073, -0.0112,
0.0724, -0.0268, 0.0140, -0.0726, 0.0298, -0.0547, -0.0182,
-0.0107, 0.0146, 0.0064, 0.0137, 0.0400, 0.0404, 0.0416,
-0.0062, -0.0203, 0.0898, 0.0027, -0.0096, -0.0264, -0.0449,
-0.0029, 0.0087, 0.0114, -0.0491, -0.0396, 0.0509, -0.0011,
-0.0017, -0.0056, 0.0091, -0.0454, -0.0321, -0.0043, 0.0226,
0.0109, -0.0096, -0.0727, -0.0142, 0.0606, 0.0411, 0.0026,
0.0196, -0.0219, -0.0235, 0.0117, 0.0440, 0.0002, -0.0376,
-0.0014, -0.0163, 0.0045, -0.0560, -0.0225, -0.0008, 0.0100,
0.0030, -0.0272, 0.1043, 0.0004, -0.0948, 0.0345, -0.0128,
-0.0462, 0.0039, -0.0171, 0.0005, 0.0370, 0.0022, -0.0059,
-0.0381, -0.0590, 0.0113, 0.0260, -0.0334, 0.0143, -0.0095,
-0.0008, -0.0221, -0.0394, 0.0033, 0.0011, -0.0254, -0.0212,
-0.0805, -0.0195, -0.0460, 0.0057, -0.0318, -0.0235, 0.0421,
-0.0037, 0.0171, 0.0647, 0.0278, -0.0105, 0.0209, 0.0293,
-0.0007, -0.0402, -0.0022, -0.0291, -0.0338, -0.0367, -0.0254,
0.0568, 0.0001, 0.0042, -0.0233, 0.0033, -0.0024, -0.0445,
0.0137, -0.0672, -0.0406, -0.0106, -0.0100, 0.0335, 0.0061,
0.0456], device='cuda:0')), ('layer2.2.bn3.running_var', tensor(1.00000e-02 *
[ 0.0595, 0.0905, 0.0489, 0.3590, 0.2604, 0.1794, 0.0293,
0.1850, 0.0837, 0.1756, 0.0821, 0.0730, 0.0829, 0.1349,
0.2706, 0.0939, 0.2051, 0.2086, 0.1292, 0.3328, 0.0349,
0.1227, 0.0433, 0.2291, 0.0904, 0.3136, 0.1490, 0.3391,
0.2423, 0.3273, 0.0440, 0.4661, 0.0698, 0.0896, 0.0853,
0.0436, 0.0596, 0.1241, 0.2758, 0.0552, 0.0990, 0.1289,
0.4335, 0.0561, 0.2638, 0.1404, 0.0658, 0.0997, 0.0960,
0.1116, 0.1063, 0.1597, 0.8768, 0.1355, 0.0914, 0.0770,
0.1618, 0.1329, 0.0333, 0.3004, 0.0567, 0.0499, 0.3102,
0.0882, 0.0918, 0.1682, 0.1115, 0.1197, 0.0751, 0.0501,
0.0949, 0.1867, 0.0613, 0.2850, 0.0295, 0.1276, 0.0321,
0.0267, 0.0827, 0.0559, 0.0318, 0.0667, 0.0451, 0.1742,
0.1945, 0.1265, 0.1100, 0.1347, 0.0704, 0.3714, 0.1511,
0.0429, 0.0386, 0.0886, 0.3287, 0.1492, 0.0752, 0.0582,
0.0571, 0.0863, 0.1531, 0.0486, 0.0172, 0.0260, 0.4604,
0.1231, 0.2298, 0.0947, 0.0870, 0.1389, 0.0920, 0.5056,
0.5739, 0.1589, 0.1181, 0.0308, 0.0613, 0.0251, 0.0720,
0.1514, 0.2761, 0.1994, 0.0919, 0.1684, 0.6196, 0.2850,
0.0207, 0.0429, 0.1474, 0.1978, 0.2132, 0.1436, 0.1357,
0.0763, 0.0313, 0.6181, 0.1072, 0.1434, 0.1977, 0.1359,
0.0629, 0.3287, 0.1180, 0.3061, 0.0835, 0.0974, 0.0628,
0.0658, 0.1003, 0.1398, 0.0446, 0.2167, 0.2343, 0.3032,
0.2004, 0.0295, 0.1439, 0.1285, 0.1461, 0.1396, 0.2423,
0.1583, 0.0265, 0.0780, 0.0354, 0.3354, 0.0325, 0.4627,
0.0609, 0.1398, 0.1869, 0.0447, 0.0532, 0.0589, 0.2163,
0.0850, 0.1508, 0.0449, 0.1526, 0.0978, 0.0727, 0.1469,
0.0427, 0.0975, 0.1134, 0.1049, 0.1863, 0.0428, 0.0808,
0.3051, 0.0507, 0.1221, 0.2744, 0.1730, 0.2590, 0.0785,
0.1229, 0.0974, 0.3107, 0.0629, 0.7663, 0.2815, 0.1194,
0.3481, 0.1382, 0.1209, 0.1409, 0.0783, 0.0652, 1.0005,
0.1223, 0.0577, 0.0777, 0.2034, 0.1629, 0.0815, 0.4480,
0.0633, 0.1148, 0.0904, 0.1012, 0.3437, 0.0766, 0.1190,
0.0195, 0.3810, 0.0768, 0.1888, 0.0991, 0.0615, 0.1441,
0.1707, 0.2569, 0.3113, 0.2091, 0.1574, 0.0584, 0.0235,
0.0268, 0.1637, 0.0846, 0.2986, 0.0610, 0.0574, 0.1921,
0.0761, 0.0326, 0.0331, 0.0820, 0.1591, 0.3313, 0.2757,
0.1228, 0.0761, 0.1775, 0.1321, 0.0119, 0.0764, 0.1803,
0.0396, 0.0972, 0.0471, 0.0563, 0.0186, 0.2829, 0.0372,
0.9353, 0.0539, 0.0247, 0.0628, 0.1704, 0.2284, 0.2899,
0.1648, 0.2671, 0.1064, 0.1567, 0.1248, 0.1167, 0.1952,
0.0651, 0.2600, 0.1006, 0.2023, 0.0940, 0.0917, 0.1908,
0.1078, 0.0889, 0.1057, 0.0693, 0.4093, 0.0840, 0.1078,
0.0594, 0.0291, 0.1699, 0.1616, 0.1116, 0.1751, 0.2053,
0.0350, 0.0984, 0.1824, 0.0538, 0.1203, 0.3626, 0.2352,
0.2133, 0.0488, 0.0742, 0.0824, 0.1280, 0.1872, 0.1359,
0.0566, 0.2297, 0.2811, 0.1095, 0.1983, 0.1370, 0.1073,
0.3433, 0.3674, 0.0277, 0.1740, 0.1188, 0.2320, 0.1189,
0.2153, 0.0325, 0.0482, 0.1188, 0.1067, 0.1222, 0.1624,
0.0662, 0.0458, 0.0479, 0.2254, 0.1904, 0.4665, 0.1420,
0.1441, 0.1362, 0.0658, 0.1848, 0.1021, 0.2926, 0.0699,
0.1113, 0.3304, 0.0899, 0.0753, 0.2197, 0.1300, 0.1116,
0.1479, 0.1070, 0.1955, 0.1032, 0.0755, 0.2223, 0.0624,
0.3153, 0.1082, 0.3106, 0.0225, 0.0904, 0.0460, 0.1110,
0.0828, 0.1732, 0.0386, 0.0396, 0.0259, 0.1351, 0.2070,
0.2754, 0.0703, 0.0245, 0.0538, 0.0498, 0.2674, 0.0788,
0.2166, 0.1832, 0.0950, 0.1452, 0.1143, 0.1060, 0.0737,
0.1230, 0.0742, 0.0433, 0.2493, 0.1209, 0.1007, 0.1244,
0.1749, 0.1615, 0.0402, 0.0401, 0.1027, 0.2070, 0.1129,
0.0619, 0.1783, 0.1768, 0.4077, 0.0166, 0.1055, 0.3412,
0.4914, 0.0650, 0.0532, 0.2688, 0.0859, 0.0936, 0.2604,
0.1346, 0.0323, 0.0469, 0.0556, 0.1422, 0.3390, 0.0757,
0.0788, 0.1035, 0.1004, 0.0868, 0.1906, 0.0885, 0.0336,
0.1154, 0.0745, 0.1149, 0.0683, 0.1970, 0.2559, 0.2099,
0.0678, 0.0662, 0.2744, 0.2926, 0.1663, 0.2072, 0.2032,
0.0343, 0.2440, 0.1676, 0.4033, 0.1863, 0.0693, 0.1847,
0.1497, 0.0961, 0.0484, 0.1674, 0.2234, 0.0478, 0.0372,
0.2079, 0.1042, 0.4142, 0.1040, 0.0838, 0.0987, 0.0559,
0.0441, 0.0962, 0.1100, 0.2354, 0.0907, 0.2922, 0.1037,
0.1191, 0.1262, 0.0511, 0.0658, 0.1069, 0.1417, 0.2075,
0.0895, 0.0374, 0.1840, 0.0580, 0.0715, 0.1032, 0.1000,
0.0810, 0.3043, 0.1522, 0.1531, 0.1320, 0.0867, 0.1022,
0.1173, 0.2301, 0.0218, 0.1425, 0.0411, 0.1608, 0.2769,
0.1652, 0.0960, 0.0540, 0.5408, 0.0820, 0.0586, 0.0606,
0.1476], device='cuda:0')), ('layer2.3.conv1.weight', tensor([[[[ 3.8050e-03]],
[[-1.9299e-02]],
[[ 1.6487e-02]],
...,
[[ 5.6285e-03]],
[[ 2.5861e-02]],
[[-4.5902e-03]]],
[[[ 2.2032e-02]],
[[-1.5341e-02]],
[[-1.6618e-03]],
...,
[[ 1.2841e-02]],
[[ 2.2107e-02]],
[[ 5.3425e-03]]],
[[[-1.1393e-02]],
[[-5.1433e-04]],
[[ 1.1396e-02]],
...,
[[ 8.2032e-03]],
[[ 4.3621e-02]],
[[ 2.3429e-03]]],
...,
[[[ 1.9219e-02]],
[[-2.8844e-02]],
[[-8.4328e-03]],
...,
[[-1.7968e-02]],
[[ 1.3575e-02]],
[[ 8.0357e-03]]],
[[[ 2.6217e-03]],
[[-6.2132e-03]],
[[-3.7966e-03]],
...,
[[-2.9308e-03]],
[[-7.1205e-03]],
[[ 9.2810e-03]]],
[[[ 8.9967e-03]],
[[-1.7275e-02]],
[[-6.6972e-03]],
...,
[[ 2.3240e-03]],
[[-1.4363e-03]],
[[-1.8595e-02]]]], device='cuda:0')), ('layer2.3.bn1.weight', tensor([ 0.1933, 0.1745, 0.2015, 0.1434, 0.1982, 0.1646, 0.1300,
0.1550, 0.1509, 0.1886, 0.1708, 0.1808, 0.1298, 0.1940,
0.1376, 0.2050, 0.1648, 0.2263, 0.2375, 0.1856, 0.1696,
0.1560, 0.1636, 0.1629, 0.1736, 0.1691, 0.1799, 0.1843,
0.1190, 0.1901, 0.1574, 0.1613, 0.1489, 0.2024, 0.1434,
0.1643, 0.1377, 0.1842, 0.1626, 0.1889, 0.1453, 0.2046,
0.1646, 0.1712, 0.1745, 0.1644, 0.1686, 0.1949, 0.1360,
0.1764, 0.1926, 0.1554, 0.1822, 0.1575, 0.1424, 0.1599,
0.1876, 0.1632, 0.2079, 0.1231, 0.1688, 0.1290, 0.2197,
0.1295, 0.1555, 0.1975, 0.1702, 0.1712, 0.2074, 0.1815,
0.1759, 0.1368, 0.1868, 0.1720, 0.1879, 0.1352, 0.1620,
0.1334, 0.1505, 0.1787, 0.1253, 0.1357, 0.1478, 0.1604,
0.1495, 0.1609, 0.1859, 0.1653, 0.1574, 0.1922, 0.1444,
0.1965, 0.1663, 0.1840, 0.1670, 0.1751, 0.1812, 0.1994,
0.2035, 0.1679, 0.1741, 0.1648, 0.1720, 0.1863, 0.1407,
0.1734, 0.1615, 0.1712, 0.1470, 0.1726, 0.1928, 0.1893,
0.1618, 0.1682, 0.2273, 0.1751, 0.2389, 0.1601, 0.1422,
0.2083, 0.2071, 0.1580, 0.1803, 0.1380, 0.1505, 0.1481,
0.1653, 0.1743], device='cuda:0')), ('layer2.3.bn1.bias', tensor([-0.1208, -0.0083, -0.1050, 0.0442, -0.0881, -0.0430, 0.0200,
0.0252, 0.0549, -0.1148, 0.0246, 0.0192, 0.0552, -0.1162,
0.1111, -0.1053, -0.0561, -0.1035, -0.1617, -0.0660, 0.0272,
0.0192, -0.0761, -0.0998, -0.0142, -0.0216, -0.0481, -0.0464,
0.0950, -0.1267, -0.0593, -0.0478, 0.0002, -0.0991, 0.0335,
0.0238, -0.0098, -0.0871, 0.0215, -0.1557, 0.0375, -0.2180,
-0.0507, 0.0625, -0.0563, -0.0467, 0.0418, -0.0344, -0.0192,
-0.0732, -0.0354, -0.0145, -0.0933, 0.0265, -0.0063, -0.0113,
-0.0038, 0.0514, -0.1111, 0.0763, -0.0559, 0.0167, -0.1314,
0.0433, 0.0502, -0.0397, -0.0432, 0.0063, -0.1221, -0.1017,
-0.0252, 0.0020, -0.1009, 0.0219, -0.1026, -0.0127, 0.0382,
0.0702, -0.0331, -0.1167, 0.1101, 0.0528, 0.1014, 0.0395,
0.0331, 0.0388, -0.0369, -0.0211, 0.0288, 0.0176, 0.0309,
-0.1166, -0.0035, -0.0111, 0.0483, -0.0758, -0.0648, -0.0809,
-0.1003, -0.0029, -0.0275, -0.0038, -0.0795, -0.1249, 0.0298,
-0.0722, 0.0292, -0.0006, 0.0562, -0.0073, 0.0240, -0.0383,
-0.0890, 0.0006, -0.1483, -0.0649, -0.2346, 0.0130, 0.0390,
-0.1281, -0.1263, -0.0202, -0.0838, 0.0609, 0.0648, 0.0477,
0.0003, -0.0583], device='cuda:0')), ('layer2.3.bn1.running_mean', tensor([-0.0822, -0.1013, 0.0489, -0.1670, -0.0105, 0.0807, 0.0209,
-0.0664, -0.0543, 0.0992, 0.0243, 0.0173, -0.0736, -0.0613,
0.1528, -0.1207, 0.0890, -0.1118, -0.1576, 0.0597, -0.0464,
-0.1529, -0.0696, 0.0264, -0.2821, -0.1644, -0.0636, 0.0041,
0.0298, 0.0946, -0.0191, 0.0336, -0.0772, -0.0575, -0.1129,
-0.1516, 0.1081, -0.0140, -0.0739, -0.0368, -0.1886, -0.1579,
0.0264, -0.1362, 0.0672, 0.0146, -0.2749, -0.0767, 0.0509,
-0.0510, -0.0273, -0.0415, 0.1404, -0.1554, -0.1184, -0.0495,
-0.0587, -0.2294, -0.1133, 0.0665, 0.0206, -0.1360, -0.0752,
-0.0915, -0.1701, -0.1609, 0.0824, -0.0262, -0.0585, 0.0262,
-0.1053, 0.0854, -0.0354, -0.1895, 0.0534, -0.1076, -0.0720,
-0.1402, 0.1141, -0.0166, 0.0162, -0.0992, -0.0633, -0.0094,
-0.1095, -0.2147, 0.0355, -0.0653, -0.2378, -0.0518, -0.0778,
0.0371, -0.0832, -0.1607, -0.1473, 0.0473, -0.2578, -0.0424,
0.0904, -0.0814, -0.0436, -0.1760, 0.1280, -0.0597, -0.2310,
0.0108, -0.0054, -0.0853, -0.1097, -0.0411, -0.2560, 0.0363,
0.1164, -0.0778, -0.0815, -0.1086, -0.0660, -0.1538, 0.0555,
-0.1939, -0.0985, -0.0590, 0.0877, -0.0549, -0.1752, -0.1228,
-0.1565, -0.0697], device='cuda:0')), ('layer2.3.bn1.running_var', tensor(1.00000e-02 *
[ 1.8803, 3.0556, 2.1960, 2.1114, 1.8599, 1.6340, 1.5958,
2.9709, 3.5660, 1.6916, 3.3489, 2.7986, 1.8298, 2.1323,
3.0437, 2.3466, 1.6651, 2.2424, 2.0913, 2.9417, 2.7975,
2.5977, 2.2954, 1.3093, 2.2524, 2.3257, 2.5828, 3.3342,
2.0726, 1.5485, 2.0778, 2.2920, 2.0533, 2.4145, 1.8949,
2.3835, 1.4622, 1.8946, 2.9274, 1.4802, 2.6932, 0.9611,
2.4207, 4.3622, 2.4407, 1.7334, 2.7640, 3.3556, 0.9961,
1.7864, 2.2049, 2.6248, 1.8303, 2.6340, 1.9024, 2.1032,
3.9532, 4.4676, 2.0111, 1.4275, 2.2896, 2.0908, 2.1290,
1.9995, 2.8074, 2.9940, 2.7579, 3.3439, 2.2246, 1.9841,
2.5353, 2.0396, 1.9245, 3.2617, 1.7174, 2.1705, 3.8627,
2.4525, 2.2035, 1.3783, 2.2477, 2.3295, 2.6514, 2.4735,
2.6446, 2.7879, 2.9613, 2.2675, 3.7311, 3.0424, 2.0382,
1.9025, 1.7612, 3.3358, 3.1243, 1.9145, 1.9225, 2.5335,
2.5224, 2.9435, 2.3916, 2.4560, 1.8098, 1.6226, 2.2661,
1.6098, 2.6710, 3.2852, 3.0790, 3.1606, 4.3522, 2.3027,
1.6371, 3.2714, 2.8349, 1.9168, 1.3235, 1.9078, 2.3744,
2.1837, 1.5258, 2.9827, 1.9452, 2.4900, 3.4940, 2.6677,
2.4700, 2.2745], device='cuda:0')), ('layer2.3.conv2.weight', tensor([[[[ 6.3425e-03, 7.8066e-03, -5.9981e-03],
[-3.4487e-03, 4.7611e-03, 3.8540e-03],
[ 2.0846e-02, 1.8216e-03, -6.1446e-04]],
[[ 1.6795e-03, 3.0943e-03, 1.6706e-02],
[ 1.7158e-03, 5.4718e-05, 5.9817e-03],
[-7.7672e-04, -5.2933e-03, 8.9279e-03]],
[[ 2.2518e-02, 3.8244e-02, 1.8780e-02],
[ 1.9846e-02, 1.4106e-02, 2.1835e-02],
[ 1.9422e-02, 1.2194e-02, 1.0690e-03]],
...,
[[ 3.5645e-02, 6.1268e-03, -1.5047e-02],
[ 3.1574e-02, -1.9045e-02, -4.1200e-02],
[ 2.8079e-02, -5.6861e-03, -3.0632e-02]],
[[ 1.9894e-02, 1.8100e-02, -2.4647e-02],
[-5.3777e-04, 1.0986e-02, -1.7435e-03],
[ 2.4066e-02, 2.0868e-02, -2.3315e-02]],
[[-2.0758e-02, 1.1063e-03, 1.4957e-02],
[-3.6696e-02, -1.4199e-02, 1.0456e-02],
[-2.8005e-02, 5.4784e-03, 3.3380e-02]]],
[[[ 2.1024e-03, 2.4975e-02, 2.7121e-02],
[ 4.1890e-03, -1.5631e-02, 4.6780e-03],
[ 1.5731e-02, -1.1325e-02, -3.7648e-02]],
[[-4.8578e-03, -1.5355e-02, 2.4391e-02],
[ 7.5786e-03, -3.4544e-02, 3.7200e-02],
[ 2.8405e-02, 2.8893e-02, 1.8144e-02]],
[[-1.7961e-02, -1.7295e-02, -3.5341e-02],
[ 5.9240e-02, -2.0096e-02, -2.5839e-02],
[ 2.0119e-02, 9.4796e-02, 4.8867e-02]],
...,
[[ 3.0112e-02, 2.4018e-02, -1.0833e-02],
[ 1.6535e-02, 1.0890e-02, 5.1681e-03],
[ 6.1969e-03, -2.8730e-02, -2.5644e-02]],
[[ 1.0316e-02, 1.1595e-02, 9.7774e-03],
[ 1.1251e-02, -2.3838e-02, 5.2018e-03],
[ 2.0050e-02, -1.5464e-02, -4.7344e-02]],
[[-2.3275e-02, -3.1601e-02, -3.3325e-02],
[-7.1235e-03, -3.8596e-02, 4.8014e-03],
[ 2.6929e-02, 2.7941e-02, 1.2472e-02]]],
[[[-2.6790e-03, -7.8069e-03, 1.5697e-02],
[-9.2817e-03, -3.1359e-02, -1.8011e-02],
[-7.4429e-03, -9.8681e-03, -4.8270e-03]],
[[ 7.6149e-03, 2.3068e-03, -3.1427e-03],
[-2.6594e-02, -5.7359e-03, 3.9425e-03],
[-1.8057e-02, -1.6741e-02, -2.0627e-02]],
[[ 2.0367e-02, -1.4038e-02, -1.4850e-02],
[ 7.0032e-03, -9.1246e-04, -6.7762e-03],
[ 1.2642e-02, 3.0074e-02, 3.2019e-02]],
...,
[[-1.3792e-02, 5.0076e-03, 9.4594e-04],
[-7.0901e-03, -3.7710e-03, -2.1320e-02],
[-1.9195e-02, 5.0308e-03, 2.0809e-02]],
[[-1.3301e-02, -1.6120e-02, -2.1279e-02],
[ 4.8702e-02, 1.5276e-02, -3.6071e-02],
[ 8.6012e-04, 6.2694e-03, 2.2975e-02]],
[[-3.5448e-03, -5.4041e-03, -2.2905e-02],
[-1.7587e-02, -6.1906e-03, 6.8526e-03],
[ 1.9518e-02, 2.4120e-02, 2.8079e-02]]],
...,
[[[-5.8243e-02, -1.9527e-03, -2.0026e-02],
[-4.2831e-02, -5.2364e-03, 2.0417e-02],
[-8.8020e-03, -6.2901e-02, -1.4732e-02]],
[[-1.2308e-02, -1.0047e-02, -9.6017e-03],
[-2.7270e-02, 3.3299e-02, -1.5759e-02],
[-1.0160e-02, 1.7439e-02, -3.3807e-03]],
[[ 3.1335e-02, -2.1841e-03, -3.8891e-03],
[ 2.1179e-03, 4.4575e-02, -8.2318e-02],
[-1.9756e-02, 1.9435e-02, -4.2395e-02]],
...,
[[ 1.8102e-02, 1.3693e-02, 6.4129e-03],
[ 2.1682e-02, -5.6668e-04, 2.9968e-02],
[-5.9797e-03, -2.3583e-02, 2.4803e-02]],
[[-1.1293e-02, -1.1813e-02, -2.6792e-02],
[ 4.6765e-03, -1.4267e-02, -1.4190e-02],
[-1.2220e-02, 1.2351e-02, 8.6050e-03]],
[[ 2.3794e-02, -3.8109e-02, 3.3384e-02],
[ 4.5388e-02, -1.8319e-03, -4.8194e-02],
[ 1.8151e-02, -1.4792e-02, -4.6150e-02]]],
[[[-7.3098e-03, 1.2843e-02, -2.0237e-02],
[ 1.2503e-03, -6.0009e-03, -8.6790e-03],
[ 1.3704e-02, 1.5641e-02, 3.3672e-03]],
[[ 5.9067e-03, 2.6577e-02, -1.8495e-02],
[-3.7595e-02, -1.3465e-02, 2.9414e-02],
[-9.6436e-04, -4.6155e-03, -5.7871e-03]],
[[ 6.6646e-02, -6.5920e-03, -3.2667e-02],
[ 4.1016e-04, 2.0832e-02, 6.2238e-02],
[-1.2086e-02, -1.4311e-02, -1.2527e-02]],
...,
[[-2.2447e-02, 7.2431e-03, -3.7030e-06],
[-3.0055e-02, -1.7382e-02, 2.8570e-03],
[ 3.5450e-03, -1.1372e-02, -3.7529e-02]],
[[ 2.5008e-02, 1.0940e-02, 1.3994e-02],
[ 2.0757e-02, 1.1284e-02, -4.4114e-02],
[-3.7229e-03, -5.7655e-03, -6.9855e-03]],
[[ 1.2280e-03, 7.1947e-03, -2.1987e-03],
[-1.3080e-02, -8.5207e-03, -1.9941e-03],
[-1.2186e-03, 5.4275e-03, 6.8117e-05]]],
[[[ 1.0657e-02, -6.6865e-03, -1.1085e-02],
[ 3.7792e-02, 4.7979e-03, -2.5554e-02],
[-1.0018e-02, -9.7269e-03, -3.1181e-02]],
[[-1.2208e-02, -1.8808e-02, -3.8413e-02],
[-4.0185e-02, -4.9522e-03, 1.8202e-02],
[-4.4533e-02, -1.2457e-02, 2.1485e-03]],
[[ 6.8635e-03, -1.0885e-02, 3.5082e-03],
[ 2.6554e-02, 8.0605e-02, 3.8870e-02],
[ 1.0307e-02, 6.1691e-02, 4.6724e-03]],
...,
[[-2.8504e-04, -3.1935e-02, -3.5198e-02],
[ 8.4391e-03, -2.8120e-02, 1.4435e-02],
[-1.0772e-02, -1.5768e-02, 1.6674e-02]],
[[-1.2566e-02, -4.5324e-03, -9.2577e-03],
[ 4.5811e-04, -9.4321e-03, 8.5911e-03],
[-9.6112e-03, -9.8607e-04, -1.4677e-02]],
[[ 1.1885e-02, -1.1211e-04, 2.4021e-03],
[-3.1185e-02, -2.3945e-02, 2.1897e-02],
[ 3.7961e-03, -2.4228e-02, -1.7371e-02]]]], device='cuda:0')), ('layer2.3.bn2.weight', tensor([ 0.2074, 0.2474, 0.1677, 0.2425, 0.2163, 0.1955, 0.2321,
0.2058, 0.2101, 0.2308, 0.2240, 0.1516, 0.1945, 0.2006,
0.2096, 0.2184, 0.1866, 0.1734, 0.1951, 0.2249, 0.2122,
0.1948, 0.2335, 0.2273, 0.2185, 0.1714, 0.1980, 0.2037,
0.1971, 0.2373, 0.1454, 0.1916, 0.2003, 0.1877, 0.2570,
0.1964, 0.1639, 0.1766, 0.1626, 0.2256, 0.2051, 0.1815,
0.2470, 0.2398, 0.2179, 0.2355, 0.2473, 0.1630, 0.1952,
0.2093, 0.1856, 0.2490, 0.2296, 0.2589, 0.2215, 0.2276,
0.1625, 0.1952, 0.1762, 0.1658, 0.2189, 0.2460, 0.1902,
0.2486, 0.2318, 0.1867, 0.2165, 0.2205, 0.2127, 0.1793,
0.1959, 0.1964, 0.1633, 0.2262, 0.2057, 0.1440, 0.2084,
0.1799, 0.1784, 0.1881, 0.1870, 0.1212, 0.1888, 0.1679,
0.1979, 0.2345, 0.1304, 0.1329, 0.2103, 0.2453, 0.1983,
0.2224, 0.1246, 0.2273, 0.1703, 0.2264, 0.1801, 0.1716,
0.2181, 0.1845, 0.1691, 0.1713, 0.1945, 0.1828, 0.2058,
0.2348, 0.2278, 0.2363, 0.2057, 0.2483, 0.2212, 0.2356,
0.2158, 0.1990, 0.2075, 0.2485, 0.2038, 0.1909, 0.1976,
0.2183, 0.1872, 0.1936, 0.1987, 0.1839, 0.2238, 0.2061,
0.1845, 0.2437], device='cuda:0')), ('layer2.3.bn2.bias', tensor([-0.0622, -0.1638, -0.0352, -0.1200, -0.0676, -0.0391, -0.1209,
-0.1094, -0.1203, -0.1041, -0.0943, 0.0809, -0.0195, -0.1124,
-0.1269, -0.0329, -0.0352, -0.0932, -0.0577, -0.0920, -0.1281,
-0.0832, -0.1282, -0.0824, -0.0811, -0.0376, -0.0664, -0.0705,
-0.0845, -0.1240, 0.0886, -0.0434, -0.0452, -0.0660, -0.1429,
-0.0634, -0.0443, -0.0639, 0.0263, -0.0819, -0.0729, -0.0702,
-0.1614, -0.1583, -0.1276, -0.1568, -0.1578, 0.0259, -0.0329,
-0.0637, -0.0773, -0.1483, -0.1030, -0.1641, -0.0940, -0.0873,
0.0424, -0.0460, -0.0348, 0.0079, -0.0671, -0.1516, -0.0784,
-0.1361, -0.0619, -0.0370, -0.0766, -0.2178, -0.0514, -0.0255,
-0.0466, -0.0802, 0.0029, -0.1203, -0.0548, 0.0834, -0.1030,
-0.0586, 0.0534, -0.0506, -0.0351, 0.1891, -0.0610, -0.0689,
-0.0343, -0.0948, 0.2578, 0.1662, -0.0801, -0.0803, -0.0505,
-0.0875, 0.2776, -0.0929, 0.0497, -0.1199, 0.0051, -0.0111,
-0.0731, -0.0604, -0.0234, -0.0865, -0.0067, 0.1228, -0.0501,
-0.1657, -0.1405, -0.1211, -0.0772, -0.1539, -0.1529, -0.1700,
-0.1135, -0.0868, -0.0574, -0.1691, -0.0490, -0.0817, -0.0362,
-0.0600, -0.0508, -0.0279, -0.0495, -0.0010, -0.1255, -0.0026,
-0.0604, -0.1522], device='cuda:0')), ('layer2.3.bn2.running_mean', tensor([-0.1549, -0.0060, -0.0605, -0.0457, 0.0092, 0.0068, -0.0444,
-0.1269, -0.0381, -0.0765, -0.0859, 0.0713, -0.0503, -0.0873,
-0.0623, -0.0661, -0.0896, -0.1515, 0.4673, -0.0772, -0.0487,
-0.0381, -0.0603, -0.0804, 0.0146, -0.0738, -0.0222, -0.0652,
-0.0429, 0.0279, -0.0947, -0.0557, -0.1171, -0.0221, -0.0400,
-0.0986, -0.1424, -0.0605, -0.0192, -0.0600, -0.1221, -0.0374,
-0.0790, -0.0305, 0.0955, 0.0140, -0.0967, -0.1262, -0.0594,
-0.0915, -0.1136, 0.0098, -0.0754, -0.0393, -0.0332, -0.0823,
0.0225, -0.0632, -0.0345, -0.0140, -0.0350, -0.0922, 0.0211,
-0.0433, -0.0399, -0.0613, -0.0440, 0.1527, -0.0116, -0.0443,
-0.0114, -0.0757, -0.0279, -0.0113, -0.0536, -0.0415, -0.0241,
-0.0548, 0.1155, -0.0196, -0.0387, -0.0828, -0.0017, 0.0087,
-0.0387, -0.0678, -0.0991, -0.0253, -0.0648, -0.1291, -0.0073,
-0.0338, -0.2040, -0.0390, 0.0705, -0.0142, -0.0301, 0.0070,
-0.0580, -0.0277, -0.0815, -0.0992, -0.1103, -0.0937, -0.0709,
-0.0963, -0.1612, -0.0992, -0.0988, -0.0414, -0.1292, -0.0926,
-0.0653, 0.0269, -0.0444, -0.0543, -0.0569, -0.0912, -0.0680,
-0.0397, -0.0364, 0.0919, -0.0683, -0.0531, -0.0474, -0.1084,
-0.0944, 0.1609], device='cuda:0')), ('layer2.3.bn2.running_var', tensor(1.00000e-02 *
[ 1.9913, 1.3043, 1.0815, 1.8623, 1.5933, 1.3955, 1.8487,
1.3133, 1.5504, 1.9491, 1.9442, 1.8743, 1.5098, 0.9817,
1.0218, 1.9433, 1.4098, 0.7613, 1.5830, 1.3695, 1.4987,
1.3357, 1.6586, 1.8053, 1.8991, 1.4500, 1.2476, 1.6056,
1.1783, 1.2233, 1.0519, 2.1000, 1.5065, 1.1556, 2.0488,
1.1154, 1.5862, 0.8787, 0.9230, 1.5492, 1.9453, 1.2760,
1.7430, 1.5253, 1.4785, 1.4867, 1.7375, 1.4603, 1.8337,
2.9399, 1.1085, 1.4974, 1.4885, 1.7417, 2.0993, 1.9032,
1.8498, 1.6129, 1.3387, 1.1766, 1.5239, 1.3989, 1.3625,
1.5993, 2.1766, 1.7906, 1.3201, 1.0197, 2.1473, 1.2366,
1.5043, 1.1989, 1.0270, 1.8937, 1.4769, 1.6563, 1.3458,
1.0507, 1.3300, 1.0955, 1.1014, 1.1372, 1.7136, 0.9505,
1.5325, 2.0421, 1.4390, 1.8833, 2.0309, 2.2243, 2.0394,
2.1292, 1.7033, 2.4443, 2.2299, 1.4546, 1.1631, 1.3103,
1.4982, 1.2010, 1.4856, 0.8767, 1.6629, 4.0979, 1.8616,
1.1214, 1.5575, 1.5589, 1.2521, 1.5151, 1.1912, 1.4333,
1.1514, 1.7364, 2.2773, 1.7117, 2.1231, 1.2076, 2.2365,
1.5001, 1.4327, 1.8832, 1.8538, 1.4616, 1.6810, 1.9788,
1.1569, 2.7033], device='cuda:0')), ('layer2.3.conv3.weight', tensor([[[[-1.8562e-03]],
[[-8.4088e-03]],
[[-2.1608e-02]],
...,
[[-1.7563e-02]],
[[-4.3702e-03]],
[[ 1.2788e-04]]],
[[[-4.0999e-03]],
[[ 2.9778e-02]],
[[ 3.3029e-02]],
...,
[[-2.0626e-02]],
[[ 4.4738e-02]],
[[-3.7762e-02]]],
[[[-5.5816e-03]],
[[ 6.2830e-03]],
[[ 4.7652e-03]],
...,
[[-3.3680e-03]],
[[-1.7510e-02]],
[[ 6.6703e-03]]],
...,
[[[-1.1869e-02]],
[[ 5.6422e-03]],
[[ 4.8070e-03]],
...,
[[ 2.1484e-03]],
[[-1.5715e-02]],
[[ 2.6373e-03]]],
[[[ 1.2345e-02]],
[[ 6.1785e-03]],
[[-4.3113e-02]],
...,
[[ 1.0104e-02]],
[[ 2.9068e-02]],
[[-8.6110e-04]]],
[[[-1.4418e-03]],
[[-1.1389e-02]],
[[-1.4606e-03]],
...,
[[-2.4021e-02]],
[[ 1.2580e-02]],
[[ 2.8908e-02]]]], device='cuda:0')), ('layer2.3.bn3.weight', tensor([ 0.0119, 0.1173, 0.0142, 0.0979, 0.0618, 0.0040, 0.0135,
0.0443, 0.1939, 0.1483, 0.1087, 0.0468, 0.0522, 0.0127,
0.0733, 0.2005, 0.0481, 0.1391, 0.1191, 0.0856, 0.1922,
0.0388, 0.0505, 0.0311, 0.1017, 0.0955, 0.1766, 0.1432,
0.0249, 0.1612, 0.0834, 0.0021, 0.0464, 0.0812, 0.0508,
0.0251, 0.0131, 0.0965, 0.0453, 0.0667, 0.1400, 0.1486,
0.1518, -0.0275, 0.0261, 0.2047, 0.0321, 0.1434, 0.1926,
0.1111, 0.0883, 0.2424, 0.1063, 0.0634, 0.0457, 0.1309,
0.1432, 0.2118, -0.0009, 0.0825, 0.0146, 0.1900, 0.1709,
0.0311, 0.2123, 0.1097, -0.0056, 0.1069, 0.2154, 0.1471,
0.0809, 0.1879, 0.1282, 0.1203, 0.0892, 0.0428, 0.0847,
0.0816, 0.1266, 0.1978, 0.0343, 0.1931, 0.0117, 0.0968,
0.1422, 0.0946, 0.1074, -0.0218, 0.0203, 0.0534, 0.0503,
0.1792, 0.1640, 0.0504, 0.0218, 0.1769, 0.0770, 0.0040,
0.0775, -0.0137, 0.0938, 0.1874, 0.0025, 0.0915, 0.0336,
0.0001, 0.1412, 0.1888, 0.0606, 0.0767, 0.0398, 0.1554,
0.0797, 0.1503, -0.0069, 0.0032, 0.1461, 0.0432, 0.0846,
0.1122, 0.0657, 0.1538, 0.0561, 0.1017, 0.0766, 0.0302,
0.0425, 0.0103, 0.1762, 0.1554, 0.1355, 0.1316, 0.1422,
-0.0002, -0.0108, 0.0273, 0.1339, 0.1425, 0.0030, 0.1100,
0.1226, 0.0744, 0.1430, 0.0569, 0.0047, 0.0774, 0.0273,
0.0945, 0.1138, 0.0591, 0.0194, 0.0361, 0.1027, 0.0124,
0.0402, 0.0051, -0.0505, 0.0306, 0.0633, 0.1056, 0.1538,
0.0071, 0.0481, -0.0294, 0.0695, 0.0070, 0.0145, 0.1263,
0.1907, 0.1059, 0.0068, 0.0592, 0.0084, -0.0011, 0.0722,
0.1757, 0.0238, 0.1174, 0.0619, 0.1449, 0.2072, 0.1723,
0.0430, 0.0194, 0.0748, 0.0358, 0.0034, 0.2697, -0.0149,
0.0609, 0.0694, 0.1492, 0.0606, 0.1092, 0.0030, 0.0927,
0.0329, 0.2003, 0.1199, 0.0014, 0.0682, 0.1515, 0.0466,
0.0442, 0.0235, 0.2072, 0.1263, 0.0426, 0.0026, 0.0922,
0.0466, 0.0758, 0.0258, 0.0309, 0.1090, 0.1266, 0.0703,
-0.0065, 0.0126, 0.0970, 0.0989, 0.1170, -0.0206, 0.2532,
0.0642, 0.1580, 0.1944, 0.0082, 0.1174, 0.0653, 0.1979,
0.0979, 0.1091, 0.0922, -0.0027, 0.1173, 0.0394, 0.0329,
0.1371, 0.0505, 0.0400, 0.0364, 0.1217, 0.0053, 0.2201,
0.1495, 0.0383, 0.0428, 0.1527, 0.0173, 0.1342, 0.0027,
0.1904, 0.1486, 0.2225, 0.1141, 0.1266, 0.1079, 0.1167,
0.0085, 0.2475, 0.0887, 0.1340, 0.0094, 0.1558, -0.0047,
0.0711, 0.0353, -0.0062, 0.0360, -0.0011, 0.1202, 0.0296,
0.0763, 0.1654, 0.0068, 0.1598, 0.1536, 0.1467, 0.0598,
0.0808, 0.0644, 0.0946, 0.0838, 0.0391, 0.0069, 0.1463,
0.1057, 0.0476, 0.2007, -0.0006, 0.0135, 0.1956, 0.1521,
0.2308, 0.0955, 0.1244, 0.0316, 0.0570, 0.2520, 0.1136,
0.0564, 0.2398, 0.1165, 0.1543, 0.1807, 0.1293, 0.1368,
0.1982, 0.0245, 0.1866, -0.0035, -0.0019, 0.0323, 0.0289,
0.0584, 0.1549, 0.0135, -0.0151, 0.0173, 0.1220, 0.1056,
0.0508, 0.0861, -0.0005, 0.0203, 0.1603, 0.1600, 0.0504,
0.0569, 0.0184, 0.0564, -0.0047, 0.0385, 0.1841, 0.0637,
0.2195, 0.0291, 0.1043, 0.0846, 0.0369, 0.0862, 0.0387,
0.1482, 0.0241, 0.0735, 0.0176, 0.0035, 0.1779, 0.0066,
0.0828, 0.0130, 0.0659, 0.0690, 0.0301, 0.0980, 0.1494,
-0.0116, 0.0948, 0.2130, 0.0883, 0.0639, 0.1180, -0.0240,
0.0549, 0.0446, 0.0556, 0.1846, 0.0257, 0.0589, 0.0477,
-0.0048, 0.0848, 0.0631, 0.0148, 0.1873, 0.1089, 0.1353,
0.1487, 0.0554, 0.1436, 0.0604, 0.0054, 0.1248, 0.0826,
0.1573, 0.0137, 0.0192, 0.0069, 0.1349, 0.1167, 0.1494,
0.0097, 0.2601, 0.0933, -0.0033, 0.0303, 0.1774, 0.1147,
0.0954, 0.0241, 0.0023, 0.1656, 0.0656, 0.0864, -0.0004,
-0.0099, 0.1035, 0.0851, 0.0574, 0.1939, 0.1244, 0.0372,
0.0412, 0.0743, 0.2109, 0.0959, 0.1913, 0.0023, 0.0807,
0.0663, 0.0390, 0.2655, 0.1125, 0.1016, 0.1639, 0.0137,
0.0239, 0.0977, 0.1802, 0.0341, 0.0643, 0.1027, 0.0243,
0.0888, 0.2672, 0.1857, 0.1321, 0.0720, 0.1274, 0.1074,
-0.0092, 0.0514, -0.0142, 0.0633, 0.1901, 0.0298, 0.1709,
0.0434, 0.2026, 0.1118, 0.1421, 0.0628, 0.0857, 0.0562,
0.0543, 0.1512, 0.1599, 0.0042, 0.0577, 0.3164, 0.1101,
0.0322, 0.0938, 0.0195, 0.0884, 0.2268, 0.0440, -0.0039,
0.1977, 0.1914, 0.0095, 0.0085, 0.0232, 0.0583, 0.1516,
0.0114, 0.1619, 0.1051, 0.1312, 0.2300, 0.1101, -0.0027,
0.1132, 0.1442, 0.1134, 0.0182, 0.0735, 0.0554, 0.0136,
0.1872, 0.0623, 0.0744, 0.0651, 0.0083, 0.0620, 0.1309,
0.0681, 0.0032, 0.0466, 0.0518, 0.0049, 0.1806, 0.0877,
0.1353, 0.0476, 0.0271, 0.0497, 0.1111, -0.0058, 0.0990,
0.0568], device='cuda:0')), ('layer2.3.bn3.bias', tensor([-1.0753e-02, 1.2296e-01, 1.8016e-02, -1.4745e-01, -1.0184e-01,
1.4736e-02, -2.6944e-03, -9.4832e-02, -7.0655e-02, -1.7460e-01,
-8.9392e-02, -6.2397e-02, 5.8483e-02, 8.1586e-03, -6.0236e-02,
-8.8126e-02, 4.0486e-02, -9.0144e-02, -6.8058e-02, -1.3243e-01,
4.6727e-03, -2.2941e-02, -9.3018e-02, 4.8685e-02, -1.6099e-01,
-1.1730e-01, -1.8592e-01, -5.0664e-02, 5.8731e-02, -4.4601e-02,
-1.0900e-01, 1.1684e-02, -6.4666e-02, -1.0929e-01, -1.0403e-01,
9.2314e-03, 1.1211e-02, -1.1996e-01, 5.7720e-02, -3.6448e-02,
-1.3682e-01, -9.1506e-02, -9.8696e-02, -1.3878e-02, -5.9053e-04,
-1.3165e-01, -2.1491e-02, -1.0817e-01, -1.4217e-01, -6.5535e-02,
-1.3517e-01, 5.4794e-02, -1.3225e-01, -7.9678e-02, -8.1386e-02,
3.1635e-03, -1.1036e-02, 7.3200e-03, 1.5462e-02, -9.3652e-02,
-2.3393e-02, 2.1940e-02, -9.7693e-02, -6.2564e-02, -7.5043e-02,
-1.2641e-01, -7.2580e-03, -9.3723e-02, 9.8195e-02, -1.1159e-01,
-1.2596e-01, -2.0165e-01, -1.3123e-01, -9.4722e-02, -6.6261e-02,
-7.9655e-02, -8.2974e-02, -1.2387e-01, 5.7409e-02, 8.0741e-02,
-9.9055e-03, -8.8603e-02, -2.6275e-02, -6.2083e-02, -9.4109e-02,
-1.4126e-01, -1.1017e-01, -4.5961e-02, -3.6952e-02, 5.6733e-02,
9.2809e-02, -3.3842e-02, 4.9556e-02, 2.5482e-02, 2.0694e-02,
1.0866e-02, -1.0336e-01, -2.4721e-02, -9.7841e-02, 1.8015e-02,
-1.1074e-01, 1.8100e-02, 3.4115e-03, -9.0010e-03, -6.0877e-02,
7.1027e-03, -1.7420e-01, -8.6625e-02, -1.4987e-02, -2.3115e-02,
-7.2371e-02, -1.7420e-01, -1.0314e-01, 5.7210e-02, 8.3013e-03,
7.0849e-06, -7.1349e-02, -6.6974e-02, -9.2546e-02, -2.9622e-02,
-4.0594e-02, 6.1131e-02, -1.0239e-01, -1.8121e-01, -1.9086e-02,
-8.2694e-02, 3.3261e-02, 2.5272e-02, -5.8061e-02, 3.7129e-02,
-9.3135e-02, -1.4024e-01, -1.6741e-01, 4.0829e-03, 1.3346e-02,
1.1912e-02, -1.5067e-01, 8.2857e-02, -7.4040e-03, -7.5757e-03,
-9.1867e-02, -1.4280e-01, -1.5459e-01, -7.5941e-02, 8.5401e-03,
-1.3061e-01, -5.0029e-02, -1.1627e-01, -1.1917e-01, -1.1245e-01,
-1.3704e-02, -6.2715e-02, -1.5910e-01, -5.0106e-02, -1.1517e-01,
2.7630e-04, -1.0772e-01, 4.6743e-02, -8.4230e-02, -4.3368e-02,
-1.6859e-01, 1.0209e-02, 3.9963e-03, -3.4581e-02, -3.6768e-02,
4.9823e-03, 1.1886e-02, -1.0845e-01, 1.1292e-03, -8.8278e-02,
1.3322e-03, -2.2974e-02, -8.4240e-03, 1.6296e-02, -1.0848e-02,
-2.6196e-02, -2.5264e-02, -1.4026e-01, -6.9171e-02, -4.4666e-02,
-1.2755e-01, -1.1186e-01, -5.6086e-02, -2.9973e-03, -1.3929e-01,
-7.8196e-02, 1.1020e-03, -1.5131e-01, 1.7389e-02, 4.5829e-02,
-1.2167e-01, -9.3535e-02, -1.3126e-01, -2.1264e-01, -1.7958e-03,
-3.6409e-02, 4.1059e-02, -3.7527e-02, -5.7838e-02, 5.7738e-03,
-7.0301e-02, -6.1492e-02, -6.4794e-02, -4.4071e-02, -2.8237e-02,
-4.0050e-02, -2.9837e-02, -2.6018e-02, -2.0245e-03, -8.6654e-02,
9.1139e-02, -8.3759e-02, 3.9878e-02, -4.8310e-02, -3.0793e-02,
-6.5844e-02, 6.1619e-02, -9.2482e-04, -8.1795e-03, -9.5234e-02,
-1.3183e-01, 7.2410e-02, -3.2863e-02, -3.7035e-02, -1.9199e-01,
-1.5521e-01, -9.5280e-02, -2.5488e-02, -1.3084e-01, -8.5780e-02,
-1.7694e-01, -1.2969e-01, -1.5030e-01, -1.0276e-01, 6.7782e-03,
-1.7357e-01, 2.8345e-02, -1.3535e-02, -1.6473e-01, -8.6814e-02,
3.3441e-02, -7.3201e-02, -1.1077e-01, -2.1831e-03, -1.5356e-01,
-3.6794e-02, -5.3170e-02, -7.1816e-02, 2.9639e-02, -6.8593e-02,
-1.8413e-01, 2.0662e-02, -2.2187e-02, -1.3400e-01, -1.9961e-01,
-1.1171e-01, -2.0527e-02, -1.0173e-01, -1.8351e-01, 9.0249e-03,
-1.3788e-01, -7.3549e-02, -6.4792e-02, 8.1229e-03, -1.4727e-01,
6.2829e-03, -8.3501e-02, -5.0290e-02, 2.6632e-02, -5.1704e-02,
2.0771e-04, -7.9790e-02, -5.8899e-02, -1.2353e-01, -1.2764e-01,
9.6215e-03, -2.3874e-02, -6.9386e-02, -1.0972e-01, -7.7112e-02,
-1.1081e-01, -8.2271e-02, -2.0116e-02, -6.6208e-02, -3.8037e-02,
3.7714e-03, -9.3259e-02, -9.8476e-02, 2.9768e-02, -3.2294e-02,
1.9760e-03, -2.1011e-02, -1.0929e-01, -1.1785e-01, -2.3031e-02,
-1.1310e-01, -1.8987e-01, 7.3373e-02, -8.3597e-02, -7.7471e-02,
4.5497e-02, -1.5749e-02, -1.2919e-01, -1.2601e-01, -3.4712e-02,
-2.2256e-02, -2.4382e-02, -7.9151e-02, -5.8132e-02, -6.3466e-02,
-6.6077e-02, 8.8458e-03, -2.0902e-02, -3.4018e-02, -5.7782e-02,
9.4693e-02, -1.5027e-01, -1.0736e-01, -9.1824e-03, -1.3116e-02,
-1.1104e-01, -1.7433e-01, 4.9677e-02, -6.9699e-02, 6.9792e-03,
-2.8829e-02, -1.6447e-01, -1.7309e-01, -1.4145e-01, 6.4827e-02,
-1.6273e-02, -7.0341e-02, -1.5254e-02, -6.1869e-02, 3.5556e-02,
-7.5051e-02, -1.7145e-01, -3.5523e-02, -3.7991e-02, -1.1560e-01,
-6.3393e-02, -1.4697e-01, -9.1959e-02, 2.9727e-02, -1.8045e-02,
-1.5468e-01, -1.1406e-02, -1.8444e-03, -7.1585e-02, 3.4634e-03,
-1.1167e-01, 2.4004e-02, -1.0770e-01, -9.2750e-02, -7.9636e-02,
-1.3726e-01, -1.2879e-01, 9.6147e-03, -2.8220e-02, -1.3269e-01,
-1.3389e-01, -5.7664e-02, -1.2113e-01, -4.0037e-02, -6.5323e-02,
-7.5729e-02, -6.8245e-03, -1.4686e-01, -5.6973e-02, -6.1606e-02,
7.2811e-02, 1.0838e-02, -1.3790e-01, -7.6565e-02, -1.8736e-02,
-1.1127e-01, -5.4532e-02, -1.5063e-01, -1.0753e-01, -7.4601e-02,
-6.4743e-02, -6.4706e-02, -1.0954e-02, -5.8167e-02, -4.2078e-02,
-1.3516e-01, -9.4269e-03, -2.3380e-02, 5.0527e-04, -1.9994e-02,
-2.4426e-02, -8.6712e-02, 1.0163e-02, -6.4395e-02, 5.7263e-02,
4.7844e-03, -5.1262e-02, -7.5730e-03, -1.2574e-01, -7.4685e-02,
4.9026e-02, 2.5039e-03, -8.6998e-02, -8.5562e-02, -1.8630e-01,
1.1889e-02, -5.6308e-03, 6.0317e-02, -7.1492e-02, 2.0359e-02,
-1.1749e-01, -1.5082e-01, -6.2635e-02, 4.7078e-02, 7.6936e-02,
-1.2096e-01, 1.4469e-02, -1.1042e-01, -2.3019e-02, 6.8692e-02,
3.3290e-02, 4.4978e-02, -4.7317e-02, 1.6809e-03, -1.8336e-01,
-4.0488e-02, 1.5075e-02, 3.0089e-02, -1.3889e-01, -3.2943e-02,
2.4125e-02, -4.8884e-02, -8.4686e-02, 4.8062e-02, 2.9681e-02,
-1.7556e-01, 2.5638e-02, -3.4004e-02, -7.7240e-02, -1.7466e-01,
2.6967e-02, 1.7453e-02, 1.4212e-02, 5.6448e-04, -4.1859e-02,
-6.3392e-02, -3.0698e-02, -1.8603e-01, -9.7148e-02, -1.1214e-01,
-1.8083e-01, -1.1350e-01, -1.2246e-01, -8.4534e-02, -7.4321e-02,
-1.8591e-01, -1.8206e-02, -8.8515e-03, -1.3595e-03, -1.1627e-01,
-1.8251e-01, -5.8070e-02, -3.9273e-03, -1.1738e-01, -4.6192e-02,
-1.2331e-01, 3.5421e-02, 7.9290e-02, -9.7586e-03, -1.4005e-01,
9.2140e-02, -1.0471e-02, 7.3859e-03, -4.6659e-02, -8.1156e-02,
-1.3107e-01, 1.7838e-02, -6.3042e-02, -1.4772e-01, -3.4397e-02,
1.1343e-01, -1.3986e-01, 3.8241e-03, -5.0460e-02, -3.3018e-03,
-1.4448e-01, 2.4820e-02, -1.2669e-01, -7.1906e-02, 8.8935e-03,
-1.0229e-02, 5.7365e-02, -1.5502e-01, -6.4526e-02, 1.2748e-02,
-7.4664e-02, -1.0345e-01, -3.0873e-02, 7.8435e-04, -2.7945e-02,
-9.3781e-02, 9.8994e-03, -8.7845e-02, -8.9566e-02, -1.3074e-01,
-5.2352e-02, -2.7542e-02, 5.4210e-02, -1.1718e-02, 5.8357e-03,
-1.1990e-01, -7.6472e-02], device='cuda:0')), ('layer2.3.bn3.running_mean', tensor([-0.0092, 0.0057, 0.0022, -0.0311, -0.0126, 0.0014, -0.0024,
-0.0111, -0.0244, -0.0212, -0.0493, -0.0028, 0.0339, 0.0052,
0.0122, -0.0013, 0.0351, -0.0162, -0.0364, -0.0051, -0.0096,
0.0016, 0.0189, 0.0100, 0.0254, 0.0084, 0.0056, -0.0698,
-0.0081, 0.0004, 0.0472, 0.0043, -0.0235, 0.0106, 0.0247,
-0.0108, -0.0061, 0.0641, -0.0490, -0.0018, -0.0128, -0.0154,
-0.0221, -0.0058, -0.0166, -0.0061, -0.0045, 0.0023, -0.0417,
0.0054, -0.0268, -0.0081, -0.0269, 0.0364, 0.0207, -0.0288,
0.0342, -0.0366, -0.0069, -0.0179, -0.0204, -0.0647, 0.0168,
-0.0018, -0.1043, 0.0188, -0.0068, 0.0190, -0.0021, -0.0099,
0.0087, -0.0153, -0.0079, 0.0125, -0.0081, 0.0251, -0.0233,
0.0116, 0.0177, -0.0252, -0.0061, 0.0227, -0.0109, -0.0023,
0.0103, 0.0167, 0.0246, 0.0023, -0.0025, -0.0308, 0.0156,
-0.0316, -0.0133, -0.0444, -0.0108, 0.0079, 0.0275, 0.0070,
0.0087, 0.0060, 0.0522, -0.0140, 0.0003, 0.0011, -0.0184,
0.0047, -0.0034, -0.0307, -0.0089, -0.0027, 0.0437, -0.0224,
-0.0036, -0.0681, 0.0076, 0.0030, -0.0192, -0.0038, -0.0035,
-0.0008, 0.0119, -0.0016, 0.0108, -0.0029, -0.0125, 0.0038,
-0.0184, 0.0089, -0.0257, -0.0192, -0.0014, 0.0053, -0.0015,
-0.0020, 0.0006, -0.0053, -0.0025, 0.0527, -0.0028, 0.0146,
0.0322, -0.0343, 0.0054, -0.0089, -0.0010, 0.0116, 0.0100,
-0.0195, 0.0362, -0.0080, 0.0020, -0.0007, -0.0035, -0.0006,
-0.0101, 0.0037, -0.0050, -0.0287, 0.0503, -0.0002, 0.0063,
0.0017, -0.0207, -0.0042, 0.0064, -0.0066, -0.0125, -0.0101,
-0.0501, -0.0234, -0.0025, -0.0023, -0.0000, 0.0049, -0.0114,
-0.0349, 0.0073, 0.0048, 0.0138, 0.0789, 0.0467, 0.0189,
0.0179, 0.0083, 0.0110, 0.0496, -0.0041, 0.0009, 0.0021,
0.0039, 0.0065, 0.0293, 0.0161, 0.0864, -0.0060, 0.0087,
0.0095, -0.0006, -0.0272, -0.0033, -0.0054, 0.0295, -0.0029,
-0.0206, -0.0048, -0.0002, 0.0140, -0.0117, -0.0033, -0.0289,
-0.0041, 0.0653, 0.0056, -0.0249, -0.0162, -0.0059, 0.0096,
0.0077, -0.0019, 0.0177, 0.0193, -0.0159, 0.0093, -0.0435,
0.0101, -0.0275, -0.0232, -0.0032, -0.0213, 0.0291, 0.0380,
0.0373, -0.0083, 0.0070, -0.0072, -0.0022, -0.0004, -0.0133,
-0.0278, 0.0183, -0.0046, -0.0192, -0.0107, -0.0017, 0.0309,
-0.0005, -0.0152, 0.0290, -0.0716, -0.0123, 0.0200, -0.0046,
-0.0380, -0.0190, 0.0005, -0.0069, -0.0265, -0.0026, 0.0158,
0.0015, -0.0149, -0.0046, -0.0274, 0.0081, -0.0070, 0.0031,
-0.0226, 0.0043, 0.0023, 0.0051, 0.0034, -0.0166, 0.0065,
0.0023, -0.0034, 0.0020, 0.0025, -0.0255, -0.0135, -0.0076,
0.0212, -0.0203, -0.0469, 0.0030, 0.0119, -0.0057, -0.0078,
0.0067, -0.0300, -0.0702, -0.0045, -0.0138, -0.0064, -0.0328,
-0.1449, -0.0205, -0.0279, -0.0049, -0.0157, -0.0828, 0.0543,
-0.0423, -0.0789, 0.0106, -0.0648, 0.0258, 0.0535, 0.0844,
-0.1167, -0.0011, -0.0239, -0.0054, 0.0098, 0.0008, -0.0118,
-0.0044, 0.0060, -0.0113, 0.0037, 0.0074, -0.0342, -0.0049,
-0.0069, -0.0218, -0.0123, -0.0097, -0.0017, -0.0061, -0.0111,
0.0032, 0.0121, -0.0102, 0.0087, -0.0179, -0.0698, -0.0088,
-0.0842, -0.0050, -0.0113, -0.0098, 0.0067, 0.0093, -0.0395,
0.0027, -0.0094, 0.0208, -0.0069, 0.0041, -0.0875, -0.0084,
0.0101, -0.0106, 0.0178, -0.0002, -0.0101, 0.0306, 0.0188,
0.0010, 0.0369, 0.0051, -0.0274, -0.0101, -0.0092, 0.0016,
-0.0291, 0.0130, -0.0128, -0.0401, 0.0469, 0.0012, -0.0092,
-0.0035, -0.0073, 0.0238, 0.0073, -0.0622, 0.0122, 0.0115,
-0.0053, 0.0027, 0.0269, -0.0012, 0.0062, 0.0287, -0.0111,
-0.0226, -0.0006, -0.0074, 0.0139, 0.0099, -0.0144, 0.0177,
-0.0083, -0.0606, 0.0060, -0.0009, 0.0029, 0.0414, 0.0317,
-0.0126, 0.0002, -0.0019, -0.0211, 0.0072, 0.0403, -0.0018,
0.0151, -0.0375, 0.0150, -0.0148, -0.0814, 0.0115, 0.0178,
0.0061, 0.0041, 0.0145, -0.0240, -0.0064, -0.0051, -0.0218,
0.0229, 0.0011, -0.1134, -0.0044, 0.0516, 0.0225, 0.0025,
0.0008, 0.0164, 0.0061, -0.0355, 0.0291, -0.0015, 0.0004,
-0.0292, -0.0964, -0.0109, -0.0473, 0.0473, -0.0192, -0.0126,
0.0050, 0.0204, -0.0007, 0.0344, -0.0209, -0.0024, -0.0132,
-0.0131, -0.0096, 0.0467, -0.0047, -0.0182, -0.0021, -0.0110,
0.0209, -0.0015, -0.0644, -0.0007, 0.0749, -0.1009, -0.0419,
0.0047, -0.0038, 0.0062, -0.0272, -0.0166, -0.0006, 0.0033,
-0.0308, -0.0187, -0.0027, -0.0018, 0.0046, -0.0233, -0.0197,
0.0165, -0.0071, -0.0200, -0.0224, -0.0854, 0.0046, 0.0050,
0.0061, -0.0509, 0.0063, 0.0021, 0.0092, -0.0147, -0.0045,
-0.0287, -0.0139, 0.0023, 0.0015, 0.0011, -0.0294, 0.0244,
0.0142, 0.0005, -0.0217, -0.0015, -0.0013, -0.0183, -0.0071,
0.0504, -0.0259, -0.0024, 0.0024, -0.0066, 0.0028, 0.0502,
-0.0089], device='cuda:0')), ('layer2.3.bn3.running_var', tensor(1.00000e-03 *
[ 0.1305, 1.7045, 0.2662, 0.6555, 0.5147, 0.1484, 0.1853,
0.4463, 1.6985, 1.4565, 0.5823, 0.3103, 0.9153, 0.2311,
0.4726, 2.3936, 1.2752, 1.3200, 1.0327, 0.9017, 2.0599,
0.4758, 0.6307, 0.5981, 1.7833, 0.7156, 1.4421, 1.7146,
0.3846, 2.4314, 0.7332, 0.2323, 0.4313, 0.5004, 0.4321,
0.4210, 0.6695, 1.6537, 0.7944, 1.2921, 1.5431, 1.6381,
1.6263, 0.2339, 0.3899, 2.1840, 0.2370, 1.2366, 2.1495,
1.0059, 0.5020, 4.3573, 1.1645, 0.7536, 0.4846, 1.7243,
1.5267, 3.1970, 0.2449, 0.7121, 0.1290, 1.5640, 1.7785,
0.1460, 1.6493, 0.8217, 0.0581, 1.4158, 3.1002, 1.3593,
0.9214, 1.8600, 1.0347, 1.1813, 1.5269, 0.5649, 0.8162,
1.3229, 1.8773, 2.5817, 0.3919, 1.7759, 0.0879, 1.0377,
1.1996, 1.1652, 1.5228, 0.2164, 0.2453, 0.7910, 0.8736,
2.0695, 1.8002, 0.4570, 0.2979, 2.0952, 0.9793, 0.1055,
0.5811, 0.1638, 0.7715, 1.8611, 0.0894, 1.0208, 0.3078,
0.2258, 1.5488, 1.9859, 1.0649, 0.8905, 0.7779, 2.5580,
0.6073, 1.8348, 0.1219, 0.3420, 1.4376, 0.3901, 1.3722,
1.0423, 0.8937, 1.3050, 0.4215, 0.7202, 0.8973, 0.3599,
0.8865, 0.2727, 1.4202, 1.7920, 1.7443, 1.1203, 0.8778,
0.1354, 0.2966, 0.3750, 1.3019, 3.0624, 0.1711, 1.3781,
1.4296, 0.7038, 1.1278, 0.4627, 0.1479, 0.6404, 0.2284,
0.8856, 1.7325, 0.5814, 0.2671, 0.3446, 1.3397, 0.1705,
0.3735, 0.2498, 0.2765, 0.5683, 0.4449, 1.1662, 1.3190,
0.3311, 0.7282, 0.1800, 1.1315, 0.1714, 0.3235, 1.3160,
2.5872, 0.6143, 0.1525, 1.2759, 0.4140, 0.3148, 0.7290,
2.6773, 0.1755, 1.4487, 0.2625, 1.9140, 1.8380, 2.0467,
0.4958, 0.2003, 0.8271, 0.6417, 0.1751, 2.1789, 0.1898,
1.0539, 0.6332, 1.5256, 0.7271, 1.6793, 0.1643, 1.3991,
0.6037, 2.7584, 1.2774, 0.1479, 0.7540, 2.5829, 0.3567,
0.3810, 0.2607, 2.5758, 1.1349, 0.3421, 0.3669, 1.0607,
1.1048, 1.3119, 1.2548, 0.2979, 1.2467, 1.2794, 1.1323,
0.3869, 0.1328, 1.2464, 0.6211, 1.3061, 0.0862, 2.8145,
0.3178, 1.7597, 2.1725, 0.1397, 1.1486, 1.3262, 2.2850,
0.9804, 1.4430, 0.7159, 0.1791, 1.0092, 0.5074, 0.5900,
2.3756, 0.5043, 0.6772, 0.3938, 1.0473, 0.1846, 1.6849,
1.7276, 0.4251, 0.7922, 1.6714, 0.3080, 1.3903, 0.3092,
2.3412, 1.5962, 1.8405, 0.9265, 1.2694, 0.8424, 1.2103,
0.2280, 1.9636, 1.1600, 1.1853, 0.2169, 1.6587, 0.2669,
0.7063, 0.1821, 0.2438, 0.2168, 0.1371, 1.1941, 0.1240,
0.7388, 1.9911, 0.1167, 2.2631, 1.2858, 1.0619, 0.6040,
1.0031, 0.5771, 1.1133, 0.8611, 0.2056, 0.1691, 1.5912,
0.9854, 0.7692, 2.4331, 0.1292, 0.1628, 1.8020, 1.5927,
2.1624, 1.2685, 1.3641, 0.6936, 0.3326, 3.8642, 1.1769,
0.5181, 1.9136, 0.8128, 1.2534, 2.1865, 1.2955, 1.8805,
1.8055, 0.1111, 1.8326, 0.1352, 0.1445, 0.2576, 0.2412,
1.1327, 1.5053, 0.1113, 0.2042, 0.1854, 1.1868, 0.7826,
0.6610, 0.6906, 0.2101, 0.1669, 1.2339, 1.2099, 0.4770,
1.2142, 0.4372, 0.7474, 0.0923, 0.2180, 1.6351, 0.6343,
2.1510, 0.2179, 0.9051, 0.6677, 0.3496, 0.8393, 0.3996,
1.3839, 0.1920, 0.8843, 0.1608, 0.1856, 1.8819, 0.3823,
0.9022, 0.2740, 0.8935, 0.6100, 0.2730, 0.9602, 0.8778,
0.2283, 1.3919, 2.3583, 0.8854, 1.2677, 0.9470, 0.1113,
0.5458, 0.6847, 0.6326, 1.6883, 0.5698, 0.5791, 1.0880,
0.1786, 1.1387, 0.8127, 0.2321, 1.9965, 0.9155, 1.6098,
1.1909, 0.3832, 2.8289, 0.8362, 0.0920, 1.2362, 1.1204,
1.3529, 0.1214, 0.2839, 0.3185, 2.3114, 1.5966, 1.7417,
0.1648, 1.9295, 1.9685, 0.1635, 0.4340, 1.9287, 0.9711,
1.0468, 0.5471, 0.1743, 3.2145, 0.7141, 0.5021, 0.1833,
0.2254, 1.1792, 0.6590, 0.8328, 2.0471, 1.2184, 0.3263,
0.7382, 1.8518, 2.3463, 0.9990, 3.0415, 0.0995, 1.6989,
0.8560, 0.7686, 2.7842, 1.1046, 0.7829, 1.4056, 0.3394,
0.6343, 0.5959, 1.9297, 0.6308, 0.8577, 1.8705, 0.4522,
1.4858, 2.7033, 2.3222, 1.6476, 0.8886, 1.0318, 1.3498,
0.1746, 0.4422, 0.1672, 0.8413, 2.8875, 0.1640, 2.3303,
0.6085, 1.7557, 0.7531, 1.3606, 0.5133, 1.1079, 0.3271,
0.3570, 1.8281, 1.8675, 0.1374, 1.2047, 3.0565, 1.4907,
0.2337, 0.5549, 0.1021, 0.9083, 2.1605, 0.6465, 0.1071,
2.2278, 2.1316, 0.1688, 0.2648, 0.1283, 0.5396, 1.1676,
0.1910, 1.4888, 1.1602, 1.8855, 3.0254, 1.2198, 0.1395,
2.0443, 1.9573, 1.2329, 0.3604, 0.7976, 0.3559, 0.2966,
2.2073, 0.7142, 0.5188, 0.9339, 0.2403, 0.4396, 1.4384,
0.6596, 0.1327, 0.6201, 0.4409, 0.2029, 2.2329, 0.7456,
1.3401, 0.3724, 0.2569, 0.7662, 1.0233, 0.2264, 1.8588,
0.6339], device='cuda:0')), ('layer3.0.conv1.weight', tensor([[[[ 1.4149e-02]],
[[ 8.5007e-03]],
[[ 1.9201e-02]],
...,
[[-4.0301e-04]],
[[-1.6396e-02]],
[[ 6.3780e-03]]],
[[[-7.8159e-03]],
[[-1.5656e-04]],
[[ 1.0262e-02]],
...,
[[ 5.9090e-03]],
[[ 8.4134e-02]],
[[-2.8860e-03]]],
[[[ 2.2822e-02]],
[[-4.3230e-02]],
[[ 7.8770e-03]],
...,
[[ 1.4022e-02]],
[[ 6.7142e-03]],
[[-3.1446e-02]]],
...,
[[[-8.5522e-05]],
[[ 8.2215e-03]],
[[-2.0979e-02]],
...,
[[-1.9268e-02]],
[[-1.8546e-02]],
[[ 3.3627e-02]]],
[[[-1.2550e-02]],
[[-5.8194e-02]],
[[ 5.3627e-03]],
...,
[[ 7.3295e-03]],
[[ 2.6446e-02]],
[[ 3.0675e-02]]],
[[[-3.8911e-03]],
[[-2.7051e-02]],
[[-2.2847e-03]],
...,
[[ 7.5297e-03]],
[[-5.2904e-02]],
[[ 6.0825e-03]]]], device='cuda:0')), ('layer3.0.bn1.weight', tensor([ 0.1946, 0.2601, 0.2029, 0.1747, 0.2215, 0.1995, 0.2640,
0.2500, 0.2342, 0.2072, 0.2172, 0.2479, 0.2861, 0.2639,
0.2448, 0.2768, 0.1862, 0.2633, 0.2423, 0.1850, 0.2633,
0.2455, 0.2833, 0.2693, 0.2729, 0.2791, 0.2744, 0.2201,
0.2304, 0.2119, 0.2682, 0.2393, 0.2065, 0.2103, 0.1802,
0.2136, 0.2190, 0.2243, 0.2371, 0.2139, 0.2046, 0.2209,
0.1994, 0.2091, 0.2018, 0.2023, 0.2911, 0.2256, 0.2194,
0.1831, 0.2852, 0.2344, 0.3128, 0.2464, 0.2380, 0.2307,
0.1979, 0.2267, 0.2337, 0.2341, 0.2802, 0.1647, 0.2803,
0.2407, 0.2042, 0.3024, 0.2114, 0.2267, 0.1883, 0.2827,
0.2215, 0.2315, 0.2330, 0.2287, 0.2992, 0.2437, 0.2398,
0.2680, 0.2352, 0.2196, 0.2743, 0.2032, 0.2107, 0.2373,
0.2054, 0.2627, 0.2570, 0.2679, 0.2402, 0.2196, 0.2359,
0.2219, 0.2065, 0.2145, 0.1937, 0.2652, 0.1711, 0.2656,
0.2181, 0.1799, 0.2528, 0.2552, 0.1657, 0.2553, 0.2602,
0.2552, 0.1868, 0.2030, 0.1685, 0.2602, 0.2258, 0.2057,
0.2460, 0.1817, 0.2335, 0.2478, 0.2260, 0.2288, 0.2435,
0.2234, 0.2226, 0.2391, 0.1664, 0.2759, 0.2846, 0.3010,
0.2254, 0.1892, 0.2445, 0.2194, 0.2523, 0.2271, 0.2227,
0.1839, 0.3044, 0.2194, 0.2243, 0.2795, 0.2475, 0.2451,
0.2267, 0.2362, 0.1989, 0.2099, 0.2307, 0.1967, 0.2383,
0.2767, 0.2448, 0.2679, 0.2379, 0.2293, 0.2976, 0.2862,
0.3137, 0.2089, 0.2370, 0.2605, 0.2073, 0.2613, 0.2536,
0.1897, 0.2381, 0.1994, 0.2133, 0.2054, 0.2239, 0.2939,
0.2420, 0.1751, 0.2585, 0.2714, 0.2666, 0.2222, 0.2338,
0.1448, 0.2367, 0.2191, 0.2261, 0.2479, 0.1759, 0.2142,
0.1761, 0.1973, 0.2791, 0.2643, 0.2255, 0.2505, 0.2510,
0.2321, 0.2263, 0.2527, 0.3157, 0.2557, 0.2281, 0.2543,
0.2325, 0.2176, 0.1627, 0.2374, 0.2041, 0.2547, 0.2005,
0.1941, 0.2861, 0.2709, 0.1976, 0.2113, 0.2182, 0.2614,
0.2776, 0.2540, 0.2102, 0.2363, 0.2581, 0.2749, 0.3118,
0.2595, 0.1762, 0.2610, 0.2945, 0.1842, 0.2027, 0.2073,
0.2360, 0.2402, 0.1846, 0.2015, 0.2306, 0.2679, 0.2340,
0.2846, 0.2296, 0.2181, 0.2513, 0.2514, 0.1995, 0.2275,
0.2484, 0.2218, 0.2879, 0.2341, 0.1724, 0.2399, 0.2207,
0.2462, 0.2465, 0.2051, 0.1957, 0.2108, 0.2571, 0.1862,
0.2053, 0.2410, 0.2301, 0.2457], device='cuda:0')), ('layer3.0.bn1.bias', tensor([-0.0716, -0.2118, -0.0776, 0.0197, -0.1609, -0.0218, -0.2534,
-0.1076, -0.1576, -0.0182, -0.1506, -0.2263, -0.1674, -0.1875,
-0.1416, -0.2482, 0.0012, -0.1473, -0.1854, -0.0933, -0.1392,
-0.1027, -0.2084, -0.1890, -0.2250, -0.2308, -0.1630, -0.1422,
-0.0097, -0.0764, -0.1343, -0.0920, -0.0534, -0.1273, 0.0199,
-0.0858, -0.1193, -0.1170, -0.1612, -0.0805, -0.0423, -0.0578,
-0.1412, -0.0036, -0.0892, -0.0374, -0.1476, -0.0248, -0.1621,
-0.0887, -0.2156, -0.0749, -0.1690, -0.0949, -0.1377, -0.1694,
-0.0707, -0.0870, -0.1093, -0.0439, -0.2401, 0.0838, -0.2995,
-0.1609, -0.0226, -0.3787, -0.1112, -0.0890, -0.0177, -0.2452,
-0.1449, -0.1050, -0.2006, -0.0787, -0.3330, -0.0827, -0.1361,
-0.2433, -0.0031, -0.1533, -0.3331, -0.0573, -0.1027, -0.1138,
-0.0437, -0.1728, -0.2564, -0.1506, -0.1109, -0.1361, -0.1181,
-0.1117, -0.1008, -0.1590, 0.0091, -0.2118, -0.0568, -0.2091,
-0.0704, -0.0842, -0.0906, -0.2344, 0.0700, -0.1869, -0.1733,
-0.2587, -0.0661, -0.0717, 0.1135, -0.1573, -0.1957, -0.0804,
-0.1211, -0.0696, -0.1198, -0.1329, -0.0089, -0.1254, -0.0360,
-0.1139, -0.0515, -0.0822, 0.0103, -0.1617, -0.1541, -0.3764,
-0.1627, 0.0025, -0.1052, -0.1416, -0.1432, -0.1132, -0.1150,
-0.0381, -0.3602, -0.1302, -0.0952, -0.1754, -0.2268, -0.1265,
-0.0841, -0.1244, -0.0576, -0.0665, -0.1428, -0.1169, -0.1089,
-0.3248, -0.1756, -0.2729, -0.1996, -0.1322, -0.3657, -0.2219,
-0.3054, -0.0775, -0.2190, -0.1458, -0.1152, -0.1375, -0.1436,
-0.0792, -0.1508, -0.0657, -0.0540, -0.1350, -0.1441, -0.1499,
-0.0971, -0.0420, -0.1492, -0.2273, -0.2078, -0.1288, -0.1164,
0.1052, -0.0969, 0.0094, -0.1471, -0.1288, 0.0119, -0.0149,
0.0104, -0.0856, -0.1134, -0.0732, -0.0971, -0.1658, -0.1940,
-0.1231, -0.1383, -0.1640, -0.2866, -0.1967, -0.2092, -0.1384,
-0.1586, -0.1384, 0.0311, -0.0995, -0.0646, -0.1402, -0.0542,
-0.0334, -0.1148, -0.2483, -0.1025, -0.1563, -0.0372, -0.1717,
-0.1893, -0.2048, -0.1561, -0.1682, -0.1151, -0.0831, -0.1567,
-0.2252, -0.0313, -0.1226, -0.2396, 0.0042, -0.1189, -0.0200,
-0.1777, -0.1839, -0.0052, -0.0108, -0.1689, -0.3388, -0.1544,
-0.2196, -0.1544, -0.0588, -0.1284, -0.3399, -0.1008, -0.1639,
-0.1698, -0.1328, -0.2455, -0.1520, -0.0480, -0.1432, -0.1242,
-0.1071, -0.1742, -0.0980, -0.0320, -0.0709, -0.1540, -0.0410,
-0.0948, -0.1573, -0.2218, -0.1688], device='cuda:0')), ('layer3.0.bn1.running_mean', tensor([-0.0399, -0.1026, -0.0971, 0.0082, -0.1664, 0.0079, -0.0677,
-0.0510, -0.1630, -0.0024, -0.1646, -0.1524, -0.2446, -0.3100,
-0.0567, -0.1291, 0.0522, -0.1162, -0.1158, -0.0804, -0.3502,
-0.1340, -0.1932, -0.1402, -0.0637, -0.1171, 0.0681, -0.0470,
-0.1764, -0.0226, -0.1165, -0.1499, -0.1169, 0.0131, -0.1215,
0.0874, -0.0499, 0.0235, -0.1106, -0.0944, -0.0040, -0.1240,
-0.0329, -0.0667, -0.0590, -0.3439, -0.1106, 0.0097, -0.1287,
-0.0399, -0.1006, -0.1439, -0.0966, -0.0507, -0.0976, -0.0495,
-0.2749, -0.2779, -0.1336, -0.1523, -0.0669, -0.1344, -0.1579,
-0.0491, -0.0385, -0.0589, -0.0940, -0.1568, -0.1891, -0.0259,
-0.0239, -0.2123, -0.2556, -0.1248, 0.1009, -0.2414, -0.1054,
-0.1133, -0.2482, -0.1906, 0.0090, -0.0878, -0.1003, -0.0385,
-0.1733, 0.0559, -0.0637, 0.0841, 0.0998, -0.0787, -0.0489,
-0.0578, 0.0981, -0.1518, -0.0482, 0.0426, -0.0866, -0.1729,
-0.1023, 0.0208, -0.1780, -0.0249, -0.1474, 0.1171, -0.2280,
0.2597, -0.1328, -0.0535, 0.0974, -0.0357, -0.0287, -0.2039,
-0.0146, -0.1005, -0.2026, 0.0939, -0.1010, -0.1421, -0.2855,
0.0787, -0.1373, -0.0069, -0.0838, -0.0422, -0.0465, -0.2355,
-0.0100, -0.0195, -0.0497, 0.0154, -0.0139, -0.0221, -0.0610,
-0.1279, -0.0678, 0.0309, -0.0797, -0.1369, -0.0916, -0.1774,
-0.0253, 0.0596, -0.2512, -0.0274, 0.1727, -0.0319, -0.2179,
0.2873, -0.1350, 0.0998, -0.0818, -0.0985, -0.1977, -0.0308,
-0.2372, -0.1539, -0.0497, -0.0836, -0.0364, -0.1514, -0.1679,
-0.1202, -0.1121, -0.1003, -0.0839, -0.0494, -0.1092, -0.1157,
-0.1176, 0.0548, -0.0435, -0.0227, -0.1024, -0.0518, -0.3975,
-0.0359, -0.0130, -0.1739, -0.0630, -0.0016, -0.1100, -0.3340,
0.0086, -0.0593, -0.1747, -0.1282, 0.0179, -0.0994, -0.0899,
-0.2281, -0.1649, -0.0796, -0.1000, -0.1624, 0.0223, -0.1690,
-0.2888, 0.0322, 0.0202, -0.0573, 0.0071, -0.1565, -0.0752,
-0.2650, -0.0406, -0.0042, -0.0357, -0.0950, 0.1302, 0.0871,
-0.0511, -0.1979, 0.0475, -0.1748, -0.1585, -0.0901, -0.0427,
-0.0693, 0.0137, -0.0125, -0.0530, -0.0735, -0.0590, -0.0604,
-0.0111, -0.0360, -0.0023, -0.2675, -0.1243, -0.1380, -0.0212,
-0.0967, 0.0710, -0.0935, -0.2304, -0.1243, -0.1149, 0.0504,
-0.2040, -0.1752, -0.1515, -0.1478, -0.0604, -0.0273, -0.1372,
-0.1142, -0.0348, -0.1766, -0.0756, -0.0734, -0.0925, -0.1436,
-0.1396, -0.0365, -0.0343, -0.1601], device='cuda:0')), ('layer3.0.bn1.running_var', tensor(1.00000e-02 *
[ 2.4749, 2.2477, 3.0931, 2.5948, 1.7693, 2.9130, 2.4502,
3.3171, 2.1401, 3.4228, 2.8781, 2.4959, 3.1959, 2.4960,
2.5712, 2.3861, 3.6448, 3.5946, 2.9332, 1.7258, 3.2366,
3.5391, 2.6394, 3.9061, 2.4495, 3.1605, 3.6886, 2.0358,
4.1725, 2.2486, 4.2039, 3.7729, 2.9680, 2.0779, 4.8116,
3.1631, 2.1006, 2.2256, 2.9065, 2.6438, 4.0002, 4.8885,
1.5860, 5.7213, 2.1787, 3.0955, 4.4179, 4.4406, 2.6441,
1.8606, 2.5386, 4.3694, 2.5426, 3.0844, 2.4394, 2.0775,
3.3568, 3.2627, 2.9782, 4.1914, 3.0073, 3.4295, 2.5119,
3.3078, 3.3674, 1.8828, 2.1843, 2.4741, 2.7811, 2.3474,
3.2555, 4.1938, 2.6081, 3.7504, 2.8055, 3.7043, 2.7055,
2.0309, 4.8638, 2.2030, 1.8394, 2.5040, 1.7486, 2.8384,
2.4695, 3.3385, 2.0290, 2.8749, 3.5751, 2.1791, 3.2850,
3.5759, 2.7566, 2.6086, 3.7130, 2.2692, 1.6783, 2.8765,
3.3771, 1.7931, 2.8719, 1.7653, 5.2274, 3.1732, 2.2128,
2.3079, 2.1104, 2.6464, 4.7872, 4.3417, 2.5776, 2.7590,
3.6631, 2.2019, 2.4447, 2.5128, 7.7180, 2.3379, 5.5302,
2.9091, 4.2429, 4.2247, 2.2463, 4.0154, 4.5467, 2.3065,
1.9610, 3.1406, 3.8615, 2.2026, 4.1997, 3.8665, 2.0208,
3.0276, 2.3197, 2.2844, 2.6556, 5.0996, 2.0209, 3.6243,
4.8061, 2.4561, 3.8334, 2.6200, 2.1135, 1.8085, 3.5808,
2.3292, 2.9455, 2.2202, 2.3129, 1.7275, 3.4701, 3.1901,
3.1335, 2.9130, 1.9680, 3.8433, 2.0795, 2.9951, 2.5078,
1.7118, 3.4449, 2.8792, 3.4597, 1.8880, 1.8088, 4.7274,
3.9279, 2.9256, 3.4635, 3.0886, 2.6878, 2.3204, 3.7233,
2.8435, 2.6824, 8.8827, 2.8475, 3.1031, 2.3916, 4.2415,
2.1642, 1.7629, 4.3786, 3.1792, 3.0811, 2.3136, 2.8617,
3.2243, 2.9157, 3.8856, 2.8837, 2.5785, 1.9481, 3.6281,
2.9699, 2.0334, 3.2474, 3.7196, 2.7125, 3.9981, 2.0597,
3.4279, 4.6253, 2.6381, 1.6661, 1.7141, 3.2129, 3.0375,
3.7680, 2.2776, 2.0483, 3.1231, 3.1748, 5.1142, 5.4999,
1.9792, 2.4391, 4.5914, 3.4677, 2.8786, 1.9715, 4.7795,
2.3308, 2.3990, 2.5072, 3.8615, 1.8054, 2.4774, 2.6237,
2.4525, 2.5018, 3.9596, 4.9305, 2.2011, 2.5555, 2.8151,
3.4213, 2.4385, 2.3883, 2.5702, 1.6848, 3.5195, 2.7375,
4.1765, 3.2937, 2.2997, 3.2040, 2.6417, 3.2747, 3.0760,
2.6207, 2.5652, 2.3334, 3.1605], device='cuda:0')), ('layer3.0.conv2.weight', tensor([[[[-3.5676e-02, -4.4159e-02, -3.1522e-02],
[-3.0261e-02, -1.0305e-02, -3.0481e-02],
[-2.9075e-02, -4.0986e-02, -4.4898e-02]],
[[-1.5381e-02, -3.0678e-02, -5.1069e-03],
[-1.4952e-02, -2.5419e-02, -1.8033e-02],
[-2.8191e-02, -2.7300e-02, -2.0191e-02]],
[[ 1.5533e-02, 2.1185e-03, 2.2384e-02],
[-1.7511e-02, -7.2976e-03, -1.2996e-02],
[-5.0300e-03, -1.6129e-02, -5.1342e-03]],
...,
[[ 4.2438e-02, 5.1855e-02, 4.6973e-02],
[ 5.1620e-02, 3.4322e-02, 5.1256e-02],
[ 3.3828e-02, 4.6830e-02, 3.3604e-02]],
[[-5.2238e-03, 1.5749e-02, -9.5059e-03],
[ 6.2110e-03, 9.6251e-03, -3.9727e-03],
[-2.9889e-03, 9.8623e-03, 1.1168e-02]],
[[-1.5140e-02, -6.9561e-03, -9.3649e-03],
[ 2.9356e-02, 1.6229e-02, 2.7333e-02],
[ 2.0968e-02, 3.8168e-02, 1.6188e-02]]],
[[[-1.1559e-02, -4.2343e-04, 2.8705e-04],
[-1.7904e-03, 6.0025e-04, -7.1231e-03],
[-2.2644e-03, -6.7109e-03, 1.1530e-02]],
[[-7.1435e-03, -7.7434e-03, -1.4644e-02],
[ 4.5507e-03, -1.0593e-02, -1.1986e-02],
[ 2.3416e-03, 1.0748e-03, -9.5772e-03]],
[[-1.1639e-02, -1.4447e-02, -8.8890e-03],
[-4.6701e-03, -9.8546e-03, -1.5753e-02],
[-1.3746e-02, -1.0631e-02, -1.1963e-02]],
...,
[[-4.2594e-03, 1.5171e-02, 8.4769e-03],
[-1.3176e-03, 1.5288e-03, 4.6005e-03],
[ 2.1575e-03, -5.1395e-04, -5.0190e-03]],
[[ 2.2688e-03, 1.5911e-03, 6.7311e-03],
[ 1.3108e-02, 2.6229e-03, 2.3430e-02],
[ 8.5807e-03, 5.5091e-03, 1.6803e-02]],
[[-7.7545e-03, -1.6997e-02, -1.0051e-02],
[-1.5485e-02, -1.4481e-02, -1.2779e-02],
[-2.8206e-02, -2.1518e-02, 7.1745e-03]]],
[[[-1.0828e-02, -3.7245e-03, 1.0759e-02],
[ 1.1467e-02, 2.3765e-02, 1.7964e-02],
[-2.0970e-03, 9.4246e-03, 1.0805e-02]],
[[ 1.2899e-02, -3.4170e-03, 5.5603e-04],
[-3.5987e-03, -1.7772e-02, -1.5287e-02],
[ 1.4051e-03, -1.5824e-02, -7.2978e-03]],
[[ 2.1011e-02, 1.4220e-02, 8.9457e-03],
[ 1.2371e-02, 8.2597e-04, -4.4019e-03],
[ 2.1452e-02, 1.1267e-02, 2.6312e-02]],
...,
[[-9.2010e-03, 1.2929e-02, 7.1729e-04],
[ 4.2680e-03, 2.8272e-02, 7.7315e-03],
[-8.5769e-03, 1.2218e-02, 7.7129e-03]],
[[-7.0493e-03, 5.1563e-03, 1.8776e-03],
[ 3.5170e-03, 3.2480e-03, -4.2209e-03],
[ 5.5588e-04, 9.9910e-03, -3.6488e-03]],
[[ 5.9354e-03, 3.2923e-03, 2.2074e-02],
[ 2.0078e-02, 3.4681e-02, 2.2642e-02],
[ 1.1302e-02, 2.8044e-02, 1.7096e-02]]],
...,
[[[-3.1205e-03, 1.4762e-02, 2.4124e-03],
[-1.4880e-02, -1.2291e-02, -8.1030e-03],
[ 2.2872e-03, -8.4939e-03, 4.0697e-03]],
[[ 2.9471e-03, 2.1305e-02, 2.1662e-02],
[-2.0600e-03, -4.2156e-03, -1.6219e-02],
[ 9.7027e-03, 1.0036e-02, 6.3643e-03]],
[[ 3.5831e-04, -5.8970e-03, -6.6091e-03],
[-8.3111e-03, -1.1051e-02, -1.7800e-02],
[-3.8104e-03, -5.1521e-03, 2.7179e-03]],
...,
[[-2.5658e-03, 9.6313e-03, 2.5836e-03],
[-2.0589e-03, -1.4066e-02, -1.0730e-02],
[-2.5331e-03, 1.0627e-02, 1.5832e-02]],
[[ 1.7151e-02, 9.3345e-04, -3.8019e-03],
[-1.0191e-02, -2.3634e-02, -1.2856e-02],
[ 2.6982e-02, 2.7909e-02, 9.9536e-03]],
[[-6.3166e-03, -1.2048e-02, 8.8745e-04],
[-1.4119e-02, -1.2392e-03, 6.1497e-03],
[ 1.9563e-03, 1.2811e-03, -1.0937e-02]]],
[[[ 1.1956e-02, 6.4232e-04, -4.1247e-03],
[ 7.4465e-03, 6.4470e-03, 3.0302e-03],
[ 2.5171e-02, 8.7522e-03, 1.7879e-03]],
[[-1.0236e-02, -2.2165e-02, -1.3870e-02],
[-1.4780e-02, -1.2104e-02, -3.0022e-02],
[-1.9075e-03, -2.5575e-02, -2.5349e-02]],
[[ 2.1705e-02, 1.8870e-02, 2.3259e-02],
[ 1.3284e-02, 1.8016e-02, 2.2417e-02],
[ 9.9931e-03, -8.8218e-03, -1.4500e-02]],
...,
[[ 9.4819e-04, 2.6234e-03, 1.5482e-03],
[ 4.4344e-03, 1.1787e-02, 1.2554e-03],
[ 5.2904e-03, 7.9525e-03, 1.6375e-02]],
[[-8.5535e-03, -1.7023e-02, -6.0318e-03],
[ 9.4575e-03, 1.5215e-02, 4.4233e-03],
[ 1.4191e-03, 1.4246e-02, 2.1175e-05]],
[[-1.2326e-02, -2.4401e-02, -1.4684e-02],
[ 1.1002e-03, 7.0458e-03, -1.0717e-02],
[ 1.8407e-02, 1.3780e-02, 1.3808e-02]]],
[[[ 9.6928e-03, 7.8074e-04, -2.2226e-03],
[ 9.0362e-03, 1.0869e-02, -1.5511e-02],
[ 1.7422e-02, 6.9707e-03, -4.4602e-03]],
[[-2.1246e-03, 1.5536e-03, -4.5620e-03],
[ 2.1761e-03, 4.7383e-03, 8.7563e-03],
[-4.0447e-03, 1.5037e-03, -2.7206e-03]],
[[ 2.7551e-03, 1.3806e-03, -1.8366e-02],
[-3.2277e-03, -4.5974e-03, 3.9618e-03],
[-1.7246e-02, -6.4586e-03, 5.6523e-03]],
...,
[[ 8.8406e-04, 7.9389e-03, 8.1300e-03],
[-3.5277e-03, 1.2971e-02, 9.3556e-03],
[-1.1059e-02, 4.4324e-03, 6.1244e-03]],
[[ 5.7130e-03, 1.1547e-02, 8.9341e-03],
[ 4.7261e-03, 1.5896e-02, 2.0979e-02],
[-1.7759e-03, 3.4242e-03, 1.6707e-02]],
[[-8.9361e-03, 1.7687e-02, 2.1593e-02],
[-3.2897e-03, -1.4576e-03, 2.7925e-02],
[-2.8845e-02, 1.7679e-03, 3.3419e-02]]]], device='cuda:0')), ('layer3.0.bn2.weight', tensor([ 0.1798, 0.2306, 0.1811, 0.1527, 0.1615, 0.1532, 0.1687,
0.1446, 0.1776, 0.1907, 0.1606, 0.2337, 0.1601, 0.1798,
0.1751, 0.1669, 0.1665, 0.1605, 0.1671, 0.1809, 0.1530,
0.1717, 0.1897, 0.1645, 0.2235, 0.1936, 0.2246, 0.2008,
0.1645, 0.1610, 0.1420, 0.1509, 0.1806, 0.1448, 0.1270,
0.2224, 0.1789, 0.2264, 0.1821, 0.1755, 0.1461, 0.2929,
0.2234, 0.1878, 0.1610, 0.1763, 0.1294, 0.1606, 0.1466,
0.2336, 0.1513, 0.2153, 0.2029, 0.1582, 0.1753, 0.1860,
0.2356, 0.1781, 0.1715, 0.1715, 0.1853, 0.1973, 0.1530,
0.1336, 0.2012, 0.1465, 0.1354, 0.1315, 0.1840, 0.2009,
0.1503, 0.2067, 0.1372, 0.1486, 0.1679, 0.2088, 0.1826,
0.1920, 0.1597, 0.1963, 0.1671, 0.2259, 0.1842, 0.1435,
0.1951, 0.3224, 0.2211, 0.1801, 0.1352, 0.1874, 0.2391,
0.1712, 0.1542, 0.1833, 0.1848, 0.1757, 0.1965, 0.2287,
0.2049, 0.1782, 0.2380, 0.1750, 0.2017, 0.1478, 0.1726,
0.2122, 0.1788, 0.1515, 0.1937, 0.1991, 0.2175, 0.1795,
0.2181, 0.1910, 0.1728, 0.1956, 0.1803, 0.1994, 0.1631,
0.1628, 0.1756, 0.1994, 0.1747, 0.1733, 0.1676, 0.1529,
0.2056, 0.1913, 0.1591, 0.1386, 0.2002, 0.1996, 0.2460,
0.1689, 0.2371, 0.2237, 0.2015, 0.1659, 0.1808, 0.1637,
0.1391, 0.1660, 0.1779, 0.1808, 0.1496, 0.1422, 0.1513,
0.1878, 0.2188, 0.1713, 0.1808, 0.1708, 0.1740, 0.1948,
0.1855, 0.1471, 0.2028, 0.1877, 0.1652, 0.2179, 0.1838,
0.1411, 0.1505, 0.1518, 0.1280, 0.1933, 0.1554, 0.2108,
0.1802, 0.1989, 0.1377, 0.1663, 0.2136, 0.1776, 0.1433,
0.2122, 0.1375, 0.1992, 0.1610, 0.2016, 0.1363, 0.1538,
0.1745, 0.1832, 0.1642, 0.1766, 0.1907, 0.1841, 0.1769,
0.1363, 0.1773, 0.2047, 0.1605, 0.1429, 0.1888, 0.1546,
0.1621, 0.2075, 0.1818, 0.2592, 0.1556, 0.1623, 0.1629,
0.1437, 0.1796, 0.2407, 0.1981, 0.1521, 0.1912, 0.1876,
0.1903, 0.2092, 0.2329, 0.1720, 0.1952, 0.2492, 0.1949,
0.1370, 0.1582, 0.1846, 0.1572, 0.1790, 0.2107, 0.1947,
0.2062, 0.2048, 0.1569, 0.1942, 0.1783, 0.1753, 0.1565,
0.2462, 0.2125, 0.2283, 0.2043, 0.1430, 0.1792, 0.1885,
0.1889, 0.1545, 0.2051, 0.1732, 0.2047, 0.1689, 0.1760,
0.2027, 0.1972, 0.1848, 0.1899, 0.1931, 0.1787, 0.1904,
0.2183, 0.2140, 0.1466, 0.1988], device='cuda:0')), ('layer3.0.bn2.bias', tensor([ 0.0773, -0.0444, -0.0011, 0.1554, -0.0344, 0.1005, 0.0093,
0.1300, 0.0225, 0.0268, 0.1820, -0.0870, 0.0875, -0.0556,
-0.0356, 0.0574, 0.0247, 0.0101, 0.1577, 0.0743, 0.2682,
0.0857, -0.0070, 0.1249, -0.0706, -0.0712, -0.0088, -0.0447,
0.0214, 0.1029, 0.2384, 0.1282, 0.0671, 0.1777, 0.1912,
-0.0200, -0.0020, -0.0530, -0.0078, -0.0101, 0.1210, -0.1773,
0.0087, -0.0088, 0.0232, 0.0687, 0.1901, 0.0318, 0.1751,
-0.0254, 0.1010, 0.0832, -0.0733, 0.1143, 0.0120, -0.0234,
-0.0493, 0.0348, -0.0171, 0.0118, -0.0069, -0.0326, 0.1289,
0.1441, 0.0088, 0.1439, 0.1763, 0.1491, 0.0492, -0.0350,
0.0852, 0.1501, 0.1618, 0.0455, -0.0269, -0.0567, 0.0437,
0.0069, 0.0741, -0.0112, 0.0835, 0.0410, 0.0302, 0.1220,
-0.0325, -0.1232, 0.0673, -0.0367, 0.1539, -0.1075, -0.0162,
0.0483, 0.1149, 0.0188, 0.0244, -0.0230, 0.0430, 0.0331,
-0.0470, -0.0014, -0.0946, 0.0166, -0.0577, 0.1232, -0.0745,
0.0574, -0.0054, 0.0449, 0.0357, 0.0356, -0.1227, -0.0197,
-0.0299, -0.0971, 0.1370, -0.1095, -0.0213, -0.0317, 0.0216,
0.0603, 0.0327, -0.0267, 0.0322, 0.0711, 0.1230, 0.0849,
-0.0542, -0.0803, 0.0505, 0.1593, 0.1004, 0.1389, -0.0184,
0.0572, -0.0356, -0.0851, -0.0102, 0.0718, -0.0122, 0.0814,
0.1117, -0.0144, -0.0080, 0.0370, 0.1115, 0.0804, 0.0444,
-0.0104, -0.0648, 0.0352, -0.0378, 0.0429, 0.0117, 0.1067,
-0.0010, 0.0942, -0.0599, -0.0109, 0.0166, -0.0481, 0.0112,
0.1343, 0.1379, 0.2243, 0.1111, -0.0404, 0.1319, -0.0412,
0.0487, -0.0235, 0.1136, 0.0616, -0.0009, 0.0025, 0.1923,
-0.0109, 0.1585, 0.0392, 0.0456, 0.0624, 0.2003, 0.1642,
-0.0360, -0.0110, 0.0634, 0.0840, -0.0357, 0.1361, 0.0075,
0.2502, 0.1275, -0.0573, 0.1072, 0.1343, -0.0718, 0.0359,
0.0222, -0.0470, 0.1273, -0.1589, 0.1498, 0.0466, 0.0935,
0.2316, -0.0205, -0.0258, -0.0233, -0.0309, -0.0345, -0.0752,
0.0002, -0.0760, -0.0305, 0.1273, 0.0917, -0.0648, 0.0051,
0.1696, 0.1308, -0.0488, 0.0752, -0.0411, -0.0846, -0.0431,
0.0355, 0.0601, 0.0774, -0.0078, 0.1132, 0.0668, 0.1066,
-0.0187, -0.0244, -0.0844, -0.0265, 0.1637, -0.0145, 0.1225,
0.0041, 0.1196, -0.0401, 0.0708, -0.0002, 0.0604, -0.0266,
-0.0951, 0.0017, -0.0708, -0.0062, -0.0031, -0.0458, 0.0098,
0.0545, -0.1338, 0.0139, -0.0439], device='cuda:0')), ('layer3.0.bn2.running_mean', tensor([ 0.0077, -0.1632, -0.0781, -0.0608, -0.0430, -0.1115, -0.0593,
0.0802, 0.0436, -0.0161, 0.2581, -0.1896, -0.0993, 0.0314,
0.0022, 0.0329, 0.0411, -0.0170, -0.0138, -0.0872, 0.0114,
-0.1517, -0.0425, -0.0144, -0.2151, -0.0621, -0.0497, -0.0544,
0.0025, 0.1105, -0.0316, 0.0159, 0.0024, -0.0415, -0.0300,
-0.0196, -0.0252, -0.0055, -0.1055, -0.0607, -0.0532, -0.2547,
-0.1226, -0.0137, -0.0267, 0.0332, -0.0221, 0.0082, 0.0403,
-0.2208, -0.0971, -0.0219, 0.0107, -0.1610, 0.1465, -0.0180,
-0.1057, -0.0286, -0.0210, 0.0245, -0.0482, -0.1282, -0.0182,
-0.0727, -0.0610, 0.0889, -0.0460, 0.1669, 0.0414, -0.1378,
-0.0747, -0.0200, 0.0570, 0.0722, -0.0562, -0.2402, -0.0797,
-0.0584, 0.0134, -0.1011, 0.0371, -0.0833, -0.0621, 0.0597,
-0.0332, -0.2799, -0.0563, -0.0916, 0.0315, -0.0758, -0.2371,
-0.0148, -0.0058, -0.0200, -0.0265, -0.0374, -0.0636, -0.0701,
-0.0214, -0.0536, -0.0711, 0.0099, 0.0330, 0.0185, 0.0223,
-0.0843, -0.0677, -0.0109, -0.2255, -0.0817, -0.1628, 0.0013,
-0.0303, -0.0124, -0.0045, -0.0362, 0.0233, -0.0631, -0.1254,
-0.1512, -0.1272, 0.0700, -0.0401, -0.0303, 0.0762, -0.0129,
-0.0711, -0.0401, -0.0438, -0.0603, -0.1206, 0.0185, -0.0570,
-0.0786, -0.2543, -0.0793, -0.0560, 0.0006, -0.0570, 0.0211,
-0.0287, -0.0901, 0.0231, 0.0068, 0.0493, 0.0070, -0.0473,
-0.1992, -0.0428, 0.0938, -0.0329, -0.0306, 0.0555, 0.1530,
-0.1836, -0.0060, -0.0939, -0.0948, -0.0893, -0.1167, 0.0336,
-0.1528, -0.0294, -0.0193, -0.0625, 0.0080, -0.1915, -0.0746,
-0.1691, -0.0013, 0.1141, -0.1205, -0.1804, -0.0300, -0.0220,
-0.1638, 0.0194, -0.0776, -0.1066, -0.1433, -0.0550, 0.1113,
-0.0417, -0.0506, -0.1196, 0.0329, -0.0353, 0.0355, 0.1399,
-0.0463, 0.0169, -0.0412, 0.0480, -0.0846, -0.0553, 0.1473,
-0.0908, -0.1179, -0.0107, -0.1951, -0.0574, 0.0319, -0.1142,
0.0128, -0.0274, 0.0021, 0.0139, -0.0170, -0.1280, -0.1051,
-0.0166, -0.1058, -0.0363, -0.0477, -0.0321, -0.2916, -0.0462,
0.0687, 0.1665, -0.0094, -0.0309, -0.0579, -0.0913, -0.0130,
-0.0942, -0.1437, -0.1334, -0.0740, 0.0240, -0.0747, -0.0402,
-0.0356, 0.0707, -0.0671, -0.1019, -0.0011, -0.0783, 0.0397,
-0.0564, -0.0315, 0.0782, -0.0274, 0.0538, -0.1581, -0.0250,
-0.0388, -0.1314, -0.0525, -0.0713, -0.1447, 0.0380, -0.0368,
0.0105, -0.1722, -0.0422, -0.0310], device='cuda:0')), ('layer3.0.bn2.running_var', tensor(1.00000e-02 *
[ 2.4971, 6.3917, 3.7757, 3.4447, 1.4011, 5.2028, 1.8794,
1.9694, 2.8385, 3.1304, 2.5858, 4.2847, 2.6033, 1.9113,
1.6202, 2.0828, 2.4818, 1.6207, 3.3265, 4.3486, 3.7262,
2.7368, 2.0821, 3.0964, 2.3398, 2.3709, 3.5501, 2.5418,
3.3347, 2.4355, 2.0456, 2.6149, 3.9866, 2.0746, 1.8190,
4.2991, 2.2773, 2.7365, 3.5927, 2.0918, 2.8687, 4.0062,
3.7560, 2.0891, 1.7922, 2.6598, 1.6791, 2.0907, 2.1429,
3.6375, 4.1841, 2.3596, 2.6301, 3.0136, 3.1813, 1.7293,
2.8389, 2.8328, 1.9990, 1.8785, 2.2518, 2.5506, 2.9145,
1.9830, 3.3494, 1.9724, 2.0222, 1.8542, 2.6416, 2.5701,
2.2786, 4.9130, 2.2091, 1.8104, 1.6407, 3.7586, 3.3614,
2.9517, 2.0158, 3.2308, 2.0438, 3.1624, 2.6094, 1.9783,
3.2787, 6.2591, 5.0946, 2.4455, 2.1123, 1.7844, 3.4620,
2.3806, 4.1135, 2.4261, 1.6276, 1.5710, 2.7871, 7.8836,
2.0267, 2.9065, 3.7048, 2.4086, 2.4271, 4.8340, 1.6128,
4.2514, 2.1922, 1.9574, 3.3226, 4.0908, 2.4342, 2.0992,
5.6624, 2.2467, 2.7752, 2.1915, 1.6753, 3.1285, 3.0080,
2.3493, 3.7280, 2.6090, 2.0048, 3.4545, 2.5546, 2.8231,
1.7872, 2.1487, 1.6275, 1.7701, 3.5818, 4.9399, 9.2070,
2.6479, 6.2077, 1.9345, 3.2082, 1.6868, 1.5822, 2.7846,
4.9797, 3.2697, 2.2366, 2.8160, 2.5459, 2.0039, 1.8081,
2.3597, 2.8656, 3.7394, 2.5460, 2.6693, 2.6066, 3.5302,
3.1243, 1.8771, 2.3008, 2.4974, 4.1455, 3.7774, 3.8217,
1.8848, 1.8778, 3.2303, 2.3978, 1.4310, 2.9679, 1.9363,
4.8435, 2.0551, 1.8466, 1.8419, 3.1398, 1.8088, 2.1971,
3.8897, 2.0487, 3.1906, 1.7753, 7.5694, 1.9416, 2.2431,
2.0017, 1.6333, 2.9487, 3.1578, 2.6862, 4.3333, 2.8618,
1.5849, 3.8602, 2.6299, 3.0534, 2.1152, 2.1253, 1.8017,
1.3211, 2.7416, 6.4856, 3.6205, 2.2004, 2.7089, 4.4885,
2.3033, 2.2613, 4.2961, 2.4033, 1.5593, 1.4839, 1.8469,
2.3687, 4.5493, 3.8645, 2.9551, 2.9946, 4.7142, 2.0369,
2.1896, 2.7691, 1.7821, 1.9321, 1.5459, 3.9600, 2.4016,
3.0047, 3.3407, 2.0319, 2.3409, 3.5333, 2.5574, 3.5102,
9.5868, 2.9372, 3.6347, 3.4225, 1.9717, 1.6813, 4.4099,
2.2028, 1.7180, 3.2634, 2.9706, 2.7053, 3.1107, 2.2743,
2.3946, 2.2372, 2.4427, 3.2171, 3.9282, 2.2310, 2.0544,
4.0094, 2.2911, 1.8881, 2.2555], device='cuda:0')), ('layer3.0.conv3.weight', tensor([[[[-9.9246e-03]],
[[ 1.9633e-02]],
[[ 3.7456e-02]],
...,
[[-7.0135e-03]],
[[-2.0328e-02]],
[[-2.6514e-02]]],
[[[-1.6656e-03]],
[[-7.8092e-04]],
[[ 1.2820e-02]],
...,
[[-1.7758e-03]],
[[-4.2129e-03]],
[[-8.0088e-03]]],
[[[-2.8890e-03]],
[[-5.8743e-03]],
[[-1.6458e-02]],
...,
[[-3.6337e-04]],
[[-2.6808e-02]],
[[ 3.0729e-02]]],
...,
[[[-9.9733e-03]],
[[-4.3565e-03]],
[[-1.4252e-02]],
...,
[[-1.0327e-02]],
[[-1.9345e-02]],
[[ 2.5460e-02]]],
[[[ 3.3327e-02]],
[[ 6.7621e-04]],
[[-9.6225e-03]],
...,
[[-1.5895e-02]],
[[-1.5170e-02]],
[[-1.9303e-02]]],
[[[-2.4183e-02]],
[[ 5.4039e-03]],
[[-2.7983e-02]],
...,
[[ 9.6453e-03]],
[[ 5.1977e-03]],
[[-1.6316e-02]]]], device='cuda:0')), ('layer3.0.bn3.weight', tensor([ 1.6133e-01, 1.1563e-01, 8.9439e-02, ..., 1.0895e-01,
1.8421e-01, 1.3261e-01], device='cuda:0')), ('layer3.0.bn3.bias', tensor([-1.4362e-02, -2.5333e-02, -2.6678e-02, ..., 1.2518e-02,
4.4802e-03, -8.8591e-03], device='cuda:0')), ('layer3.0.bn3.running_mean', tensor([-2.6332e-02, -3.5791e-02, -3.0687e-02, ..., -6.8786e-02,
-4.4042e-02, 4.1869e-02], device='cuda:0')), ('layer3.0.bn3.running_var', tensor([ 5.1348e-03, 1.6210e-03, 1.1303e-03, ..., 1.6699e-03,
5.3793e-03, 4.6198e-03], device='cuda:0')), ('layer3.0.downsample.0.weight', tensor([[[[ 2.8775e-02]],
[[ 2.0881e-02]],
[[ 8.0852e-03]],
...,
[[ 1.5601e-02]],
[[ 4.1752e-02]],
[[-2.3844e-02]]],
[[[-9.3443e-03]],
[[ 3.6557e-03]],
[[-1.2121e-02]],
...,
[[ 1.2917e-03]],
[[-1.7267e-02]],
[[-3.3483e-03]]],
[[[ 1.7818e-02]],
[[-1.5537e-02]],
[[-2.0420e-03]],
...,
[[-1.3075e-02]],
[[-8.0338e-03]],
[[ 1.0759e-03]]],
...,
[[[-3.4276e-03]],
[[-1.3855e-02]],
[[-1.1908e-02]],
...,
[[-4.2695e-03]],
[[ 8.3579e-03]],
[[ 5.0778e-02]]],
[[[ 2.5238e-02]],
[[-1.8766e-02]],
[[-1.3570e-02]],
...,
[[ 5.9170e-03]],
[[ 2.2531e-02]],
[[ 5.9379e-02]]],
[[[-3.9945e-02]],
[[-1.7317e-02]],
[[ 4.6550e-02]],
...,
[[ 7.9202e-03]],
[[ 1.0088e-02]],
[[ 1.4772e-02]]]], device='cuda:0')), ('layer3.0.downsample.1.weight', tensor([ 1.0480e-01, 9.4626e-02, 6.6850e-02, ..., 9.3046e-02,
8.4846e-02, 1.7689e-01], device='cuda:0')), ('layer3.0.downsample.1.bias', tensor([-1.4362e-02, -2.5333e-02, -2.6678e-02, ..., 1.2518e-02,
4.4802e-03, -8.8591e-03], device='cuda:0')), ('layer3.0.downsample.1.running_mean', tensor([ 3.5735e-02, 1.9263e-02, -4.6170e-02, ..., -1.4835e-01,
4.2501e-02, 1.7570e-02], device='cuda:0')), ('layer3.0.downsample.1.running_var', tensor([ 1.0624e-02, 3.9305e-03, 2.4064e-03, ..., 4.6739e-03,
7.6096e-03, 2.0152e-02], device='cuda:0')), ('layer3.1.conv1.weight', tensor([[[[ 3.9551e-03]],
[[-1.3039e-02]],
[[ 1.5156e-03]],
...,
[[ 1.3798e-02]],
[[-8.8331e-03]],
[[-1.8208e-03]]],
[[[ 3.8385e-03]],
[[-3.0152e-02]],
[[-9.4545e-04]],
...,
[[-1.1199e-02]],
[[-1.2743e-02]],
[[-1.2080e-02]]],
[[[-4.4737e-03]],
[[ 6.9822e-03]],
[[ 4.8189e-03]],
...,
[[-8.4677e-03]],
[[-6.3884e-03]],
[[-1.3627e-02]]],
...,
[[[-1.4229e-02]],
[[-7.7992e-03]],
[[-6.5336e-03]],
...,
[[ 1.8012e-02]],
[[-1.0390e-03]],
[[-3.7104e-03]]],
[[[-3.9599e-03]],
[[-2.9714e-03]],
[[-2.7192e-03]],
...,
[[-3.1509e-03]],
[[-1.2503e-02]],
[[ 3.1340e-02]]],
[[[ 8.8526e-03]],
[[ 4.2982e-03]],
[[ 3.9464e-03]],
...,
[[-6.7523e-03]],
[[-2.6792e-03]],
[[ 1.1787e-03]]]], device='cuda:0')), ('layer3.1.bn1.weight', tensor([ 0.1272, 0.1082, 0.1116, 0.1127, 0.1398, 0.1230, 0.1527,
0.1691, 0.1548, 0.0954, 0.1592, 0.1576, 0.1536, 0.1489,
0.1296, 0.1624, 0.2288, 0.1425, 0.1697, 0.1377, 0.1467,
0.1772, 0.1407, 0.1456, 0.1365, 0.1424, 0.1050, 0.1253,
0.1927, 0.1208, 0.1184, 0.1757, 0.1292, 0.1253, 0.1822,
0.1592, 0.1506, 0.1569, 0.1689, 0.1787, 0.1732, 0.1845,
0.1267, 0.1831, 0.1693, 0.1621, 0.1526, 0.1276, 0.1446,
0.1489, 0.1491, 0.1670, 0.1531, 0.1349, 0.1365, 0.1655,
0.1940, 0.1671, 0.1559, 0.1311, 0.2199, 0.2139, 0.1894,
0.1218, 0.1321, 0.1719, 0.1226, 0.1500, 0.1235, 0.1654,
0.1364, 0.1341, 0.1682, 0.1612, 0.1420, 0.1354, 0.1986,
0.1501, 0.1856, 0.2027, 0.1599, 0.1339, 0.2498, 0.1636,
0.1874, 0.1955, 0.1543, 0.1632, 0.1347, 0.1574, 0.2249,
0.1630, 0.1435, 0.1655, 0.1928, 0.1806, 0.0972, 0.1760,
0.1727, 0.1930, 0.1423, 0.1359, 0.1457, 0.2264, 0.1311,
0.1175, 0.1465, 0.1911, 0.2265, 0.1366, 0.1760, 0.2234,
0.1354, 0.1680, 0.1412, 0.1940, 0.1522, 0.2007, 0.1365,
0.1594, 0.1409, 0.1560, 0.1449, 0.1585, 0.1363, 0.1593,
0.1610, 0.1640, 0.1203, 0.1814, 0.1700, 0.1435, 0.1512,
0.1339, 0.1615, 0.1384, 0.1730, 0.1316, 0.1534, 0.1654,
0.1890, 0.1417, 0.1257, 0.1474, 0.1626, 0.1308, 0.1961,
0.1080, 0.1513, 0.1742, 0.1658, 0.1728, 0.1874, 0.1260,
0.1297, 0.1786, 0.1943, 0.1530, 0.1568, 0.1227, 0.1862,
0.1422, 0.1873, 0.1863, 0.1619, 0.1608, 0.1859, 0.1504,
0.2033, 0.1943, 0.1557, 0.2199, 0.1524, 0.1532, 0.1425,
0.1613, 0.1259, 0.1423, 0.1486, 0.1405, 0.1417, 0.1554,
0.1563, 0.1198, 0.1068, 0.1426, 0.1888, 0.1763, 0.1800,
0.1325, 0.1631, 0.1777, 0.1611, 0.1331, 0.1211, 0.1066,
0.1448, 0.1730, 0.1492, 0.1555, 0.1708, 0.1653, 0.1348,
0.1549, 0.3033, 0.1381, 0.1261, 0.1619, 0.1195, 0.1495,
0.1230, 0.1882, 0.1438, 0.1168, 0.1918, 0.1818, 0.1560,
0.1437, 0.1787, 0.1085, 0.1609, 0.1572, 0.1968, 0.1669,
0.1504, 0.2114, 0.1288, 0.1489, 0.1635, 0.1412, 0.1655,
0.1488, 0.1615, 0.1538, 0.1998, 0.1970, 0.1595, 0.1608,
0.1288, 0.1136, 0.1276, 0.1462, 0.1568, 0.1774, 0.1335,
0.1226, 0.1640, 0.2121, 0.1454, 0.1803, 0.1180, 0.1704,
0.1672, 0.1606, 0.1388, 0.1537], device='cuda:0')), ('layer3.1.bn1.bias', tensor([ 0.1220, 0.0849, 0.0442, 0.0460, -0.0208, 0.0386, -0.0013,
-0.0261, -0.0440, 0.1195, 0.0265, -0.1411, 0.0120, -0.0370,
0.0165, -0.0478, -0.1123, -0.0109, -0.0267, -0.0049, -0.0377,
-0.0759, 0.0586, 0.0069, 0.0698, -0.0294, 0.1259, 0.0467,
-0.0879, 0.0762, 0.0085, -0.0214, 0.0181, 0.0621, -0.0463,
-0.0111, -0.0033, 0.0227, -0.0001, -0.0231, -0.0032, -0.0186,
0.0417, -0.0588, -0.0854, -0.0733, 0.0279, 0.0692, 0.0105,
0.0680, -0.0148, 0.0032, 0.0212, 0.0705, -0.0056, -0.0139,
-0.0966, -0.0424, 0.0265, 0.0201, -0.1035, -0.1270, -0.0084,
0.0102, -0.0042, -0.0963, 0.0177, 0.0248, -0.0235, -0.0568,
0.0188, 0.0118, -0.0356, -0.0626, -0.0140, -0.0008, -0.1244,
-0.0798, -0.0727, -0.1069, -0.0327, -0.0002, -0.0998, -0.0035,
-0.0289, -0.1033, -0.0064, -0.0954, -0.0065, 0.0204, -0.1070,
-0.0471, -0.0214, -0.0733, -0.0438, -0.0614, 0.0871, -0.0884,
-0.0560, -0.1042, 0.0648, -0.0031, -0.0457, -0.1412, 0.0462,
0.0481, 0.0017, -0.0226, -0.1208, 0.0036, -0.0413, -0.0780,
0.0093, -0.0410, -0.0152, -0.0845, 0.0667, -0.0947, -0.0061,
-0.0189, 0.0402, -0.0470, -0.0234, -0.0337, 0.0383, 0.0031,
-0.0054, -0.0378, 0.0299, -0.1009, -0.0339, -0.0025, -0.0001,
0.0298, -0.0078, 0.0273, -0.0539, 0.0369, -0.0140, 0.0068,
-0.0454, 0.0312, 0.0116, 0.0357, -0.0164, 0.1747, -0.0539,
0.0508, 0.0657, -0.0289, -0.0171, -0.0593, -0.1397, 0.0482,
0.0834, -0.0039, -0.0679, 0.0024, -0.0332, 0.0445, -0.1076,
-0.0029, -0.0289, -0.0701, -0.0616, -0.0060, -0.0572, -0.0224,
-0.1307, -0.0643, 0.0093, -0.1303, -0.0412, -0.0119, -0.0028,
0.0171, 0.0652, 0.1337, 0.0225, -0.0117, 0.0160, -0.0572,
0.0256, 0.0753, 0.0293, 0.0021, -0.0720, -0.0424, -0.1075,
0.0388, -0.0436, -0.0470, -0.0393, 0.0345, 0.0537, 0.0608,
0.0419, -0.0454, 0.0227, -0.0682, -0.0075, -0.0520, 0.0636,
0.0231, -0.1542, 0.0548, 0.0749, -0.0156, 0.1031, 0.0514,
0.0645, -0.1241, 0.0237, 0.0170, -0.0505, -0.0809, -0.0303,
-0.0425, -0.0454, 0.1494, -0.0689, -0.0385, -0.0466, -0.0403,
0.0466, -0.1054, 0.0470, -0.0253, -0.0310, -0.0036, -0.0601,
-0.0155, 0.0056, -0.0120, -0.1160, -0.0940, -0.0471, -0.0230,
-0.0016, 0.0907, 0.0355, 0.0026, -0.0536, -0.0204, 0.0107,
0.0135, 0.0067, -0.1120, -0.0177, -0.0384, 0.0662, -0.0432,
-0.0568, -0.0502, 0.0091, 0.0111], device='cuda:0')), ('layer3.1.bn1.running_mean', tensor([-0.1485, -0.1910, -0.0903, -0.0660, -0.1012, -0.0185, -0.2125,
0.0729, -0.0914, -0.2705, -0.1277, -0.1522, -0.1728, -0.1039,
-0.0929, 0.0684, -0.2347, 0.0000, 0.0503, -0.0755, -0.2032,
-0.0672, 0.0002, -0.0546, 0.5582, -0.0808, 0.0472, 0.0159,
-0.0617, -0.0717, -0.1309, -0.0450, -0.0219, -0.0514, 0.1454,
-0.0421, -0.0474, -0.0101, 0.0598, 0.0456, -0.0266, -0.0133,
-0.1201, 0.2599, 0.0651, -0.0487, -0.0316, 0.0771, 0.1701,
-0.0666, 0.0597, -0.1273, 0.0048, -0.0968, 0.0938, -0.1387,
-0.0890, -0.1125, -0.0473, 0.0697, -0.1245, -0.0552, -0.4373,
-0.3141, -0.1498, -0.0918, -0.0057, 0.0181, -0.2194, -0.0709,
0.0723, -0.0767, -0.0599, -0.0521, -0.1189, -0.0675, -0.1187,
-0.0915, -0.1090, 0.0425, -0.0274, 0.0596, -0.1960, -0.1017,
-0.0794, -0.1178, -0.0271, -0.0244, -0.1185, -0.0109, 0.0500,
-0.0484, 0.0225, -0.0953, -0.0461, -0.0410, 0.0105, -0.0349,
-0.0023, -0.2003, -0.0037, -0.0626, 0.0169, 0.0752, -0.0644,
0.0362, -0.3832, -0.0512, -0.0559, 0.0394, 0.2043, -0.1200,
-0.0849, -0.0455, -0.0348, -0.1095, 0.0797, -0.0062, 0.0396,
-0.1377, -0.0762, -0.0427, -0.1727, 0.2079, -0.0358, -0.0205,
-0.0715, -0.0437, -0.2081, -0.0930, 0.0401, -0.1776, -0.0126,
0.0601, -0.1357, 0.2286, -0.2661, -0.1120, -0.1474, -0.0391,
-0.0502, -0.1812, -0.0553, 0.0428, 0.0710, -0.8818, -0.0355,
-0.0277, -0.3925, -0.0263, -0.1105, -0.0343, -0.0732, 0.1143,
0.0104, -0.0561, 0.0396, -0.0380, -0.0086, -0.2721, -0.0420,
-0.0612, -0.2045, 0.0575, -0.0693, -0.0437, 0.0115, -0.2726,
-0.0730, 0.1928, -0.0457, -0.0957, 0.1624, -0.1692, -0.0452,
-0.0262, 0.0766, 0.1045, -0.0113, -0.0207, -0.0310, -0.0152,
-0.0111, -0.1776, -0.0455, 0.1424, -0.0428, 0.0034, -0.1280,
-0.0283, 0.0567, 0.0146, -0.0482, 0.0834, -0.1673, -0.1723,
0.0855, -0.0574, 0.0057, 0.0649, 0.0209, -0.0052, -0.1115,
0.1325, 0.0822, -0.1792, 0.0641, -0.0965, -0.0499, 0.0171,
0.0737, 0.0441, -0.0273, -0.1251, 0.1324, -0.0715, -0.0300,
-0.0342, 0.0635, -0.4435, 0.0088, -0.0803, 0.0740, -0.0160,
0.0074, 0.1227, -0.7381, -0.0973, 0.0363, -0.0193, -0.0309,
-0.1205, -0.1236, -0.0313, -0.0407, -0.0481, -0.0295, 0.0466,
-0.0320, -0.1188, 0.0529, 0.0320, -0.1656, 0.1338, 0.1201,
-0.0923, -0.1586, -0.1075, -0.0789, 0.0647, 0.1004, 0.0191,
0.0666, -0.0897, -0.0239, -0.0233], device='cuda:0')), ('layer3.1.bn1.running_var', tensor([ 0.0299, 0.0279, 0.0176, 0.0203, 0.0107, 0.0161, 0.0326,
0.0613, 0.0164, 0.0184, 0.0268, 0.0114, 0.0282, 0.0185,
0.0174, 0.0136, 0.0340, 0.0231, 0.0168, 0.0252, 0.0176,
0.0217, 0.0491, 0.0228, 0.0257, 0.0182, 0.0221, 0.0184,
0.0155, 0.0423, 0.0148, 0.0422, 0.0242, 0.0356, 0.0164,
0.0197, 0.0242, 0.0284, 0.0225, 0.0302, 0.0196, 0.0308,
0.0170, 0.0313, 0.0155, 0.0187, 0.0187, 0.0170, 0.0299,
0.0273, 0.0226, 0.0187, 0.0197, 0.0211, 0.0143, 0.0202,
0.0195, 0.0206, 0.0247, 0.0159, 0.0181, 0.0139, 0.1146,
0.0167, 0.0234, 0.0128, 0.0188, 0.0353, 0.0124, 0.0186,
0.0119, 0.0199, 0.0165, 0.0109, 0.0152, 0.0172, 0.0135,
0.0129, 0.0164, 0.0253, 0.0226, 0.0192, 0.0245, 0.0220,
0.0176, 0.0287, 0.0150, 0.0135, 0.0258, 0.0343, 0.0180,
0.0159, 0.0170, 0.0118, 0.0147, 0.0230, 0.0133, 0.0166,
0.0117, 0.0165, 0.0333, 0.0159, 0.0123, 0.0255, 0.0133,
0.0154, 0.0275, 0.0340, 0.0205, 0.0138, 0.0157, 0.0344,
0.0157, 0.0145, 0.0212, 0.0266, 0.0340, 0.0191, 0.0145,
0.0256, 0.0343, 0.0170, 0.0211, 0.0201, 0.0411, 0.0204,
0.0339, 0.0250, 0.0281, 0.0213, 0.0188, 0.0228, 0.0254,
0.0212, 0.0376, 0.0224, 0.0167, 0.0154, 0.0214, 0.0262,
0.0290, 0.0294, 0.0153, 0.0240, 0.0234, 0.0460, 0.0332,
0.0230, 0.0465, 0.0364, 0.0216, 0.0125, 0.0156, 0.0127,
0.0188, 0.0266, 0.0157, 0.0242, 0.0159, 0.0177, 0.0157,
0.0175, 0.0193, 0.0123, 0.0195, 0.0127, 0.0206, 0.0175,
0.0156, 0.0200, 0.0301, 0.0191, 0.0224, 0.0174, 0.0155,
0.0264, 0.0200, 0.0310, 0.0256, 0.0120, 0.0137, 0.0214,
0.0199, 0.0191, 0.0161, 0.0259, 0.0190, 0.0159, 0.0155,
0.0178, 0.0203, 0.0177, 0.0152, 0.0153, 0.0263, 0.0181,
0.0204, 0.0159, 0.0300, 0.0142, 0.0192, 0.0181, 0.0314,
0.0278, 0.0722, 0.0253, 0.0364, 0.0289, 0.0282, 0.0235,
0.0196, 0.0191, 0.0283, 0.0277, 0.0164, 0.0125, 0.0197,
0.0166, 0.0173, 0.0252, 0.0113, 0.0135, 0.0218, 0.0143,
0.0330, 0.0208, 0.0329, 0.0161, 0.0125, 0.0164, 0.0113,
0.0222, 0.0171, 0.0162, 0.0216, 0.0169, 0.0117, 0.0221,
0.0182, 0.0163, 0.0205, 0.0237, 0.0145, 0.0158, 0.0219,
0.0210, 0.0287, 0.0128, 0.0249, 0.0175, 0.0135, 0.0154,
0.0197, 0.0177, 0.0158, 0.0278], device='cuda:0')), ('layer3.1.conv2.weight', tensor([[[[-2.2209e-03, 2.3680e-03, -9.0066e-03],
[ 3.2583e-03, -1.5061e-03, 5.6289e-03],
[ 1.4029e-03, -1.7712e-04, -2.0088e-03]],
[[ 2.6363e-02, 3.1852e-02, 8.0712e-03],
[ 5.2496e-03, 1.5645e-03, -6.0794e-03],
[-9.5910e-03, -2.8320e-02, -1.0562e-02]],
[[-4.6022e-03, 4.7448e-03, -1.1191e-03],
[-1.2140e-02, 7.7773e-04, -1.7544e-03],
[ 1.5518e-02, 1.3508e-02, 1.2042e-02]],
...,
[[ 1.2558e-02, 6.9358e-03, 4.4854e-03],
[-1.4017e-02, -4.9447e-03, 1.1854e-02],
[ 1.2114e-02, 1.4067e-02, 1.1647e-02]],
[[ 4.3703e-03, 7.9354e-03, 4.4662e-03],
[ 1.8583e-03, 6.6458e-03, 3.0552e-03],
[ 1.7924e-03, 3.8443e-03, -4.4407e-03]],
[[-2.9302e-02, -2.7518e-02, -1.3885e-02],
[-1.4232e-02, -1.6384e-03, -1.3593e-04],
[ 9.8944e-03, 2.4967e-02, 9.9636e-03]]],
[[[ 1.1274e-03, 1.9467e-03, -4.3517e-03],
[ 8.5271e-03, 3.2832e-03, 6.5055e-04],
[-2.2252e-02, -2.9143e-02, -1.4377e-02]],
[[-7.6910e-04, 1.3005e-02, -1.3075e-02],
[-8.2980e-03, 9.6980e-03, -7.3141e-03],
[-1.3871e-03, 2.9279e-03, 8.6375e-03]],
[[-8.5891e-03, -1.7072e-02, -1.3134e-02],
[ 9.6039e-03, 1.1632e-02, -1.0393e-02],
[ 6.5844e-03, 1.6355e-02, 6.8924e-03]],
...,
[[ 7.2898e-03, -2.5705e-02, -5.8938e-03],
[ 1.0163e-02, 1.7513e-03, 2.4627e-03],
[ 1.3991e-02, 2.0732e-02, 1.6026e-02]],
[[ 2.6139e-03, 4.8715e-03, 5.5723e-03],
[ 9.5424e-03, 5.0184e-03, 1.3310e-02],
[-9.0100e-03, -1.8758e-03, 4.6210e-03]],
[[-1.0261e-02, -1.5864e-02, -3.7571e-02],
[-2.8334e-02, -4.4040e-02, -5.0564e-02],
[-4.4633e-02, -7.7627e-02, -6.3077e-02]]],
[[[-8.0014e-03, -3.2197e-02, -3.3246e-02],
[ 1.4146e-02, 6.8184e-03, 3.4801e-03],
[ 5.5120e-02, 6.0446e-02, 5.1732e-02]],
[[ 2.0362e-04, 1.8760e-02, 1.0876e-02],
[-8.8109e-03, 1.2179e-02, 1.3240e-02],
[-1.0618e-02, -6.2681e-03, -1.7830e-03]],
[[ 1.7463e-02, 9.1279e-03, 6.0448e-03],
[ 9.0012e-03, 9.2734e-03, 3.0767e-03],
[-4.4393e-03, -1.9744e-02, -1.1693e-02]],
...,
[[ 1.9538e-02, 3.8320e-02, 3.2642e-02],
[-9.7889e-03, 1.8268e-03, 7.8323e-03],
[-3.1404e-02, -3.0507e-02, -1.4063e-02]],
[[ 1.7386e-03, 4.6947e-03, 1.3606e-02],
[-3.8464e-03, -9.0910e-04, 7.1564e-03],
[-1.2988e-02, -3.1593e-03, -5.1058e-03]],
[[-3.9896e-03, -1.3200e-02, -1.2730e-02],
[ 8.8646e-03, -2.0722e-03, -2.9879e-03],
[ 1.6947e-03, -3.6365e-03, 5.4581e-03]]],
...,
[[[ 1.1026e-03, 7.2603e-04, 1.1215e-03],
[ 1.4615e-02, 1.5489e-02, 1.1199e-02],
[ 5.5280e-03, 1.8136e-02, 1.3456e-02]],
[[-2.3531e-03, -1.7455e-03, 1.1275e-03],
[-9.8037e-03, -1.4158e-02, -2.5560e-03],
[-2.8319e-04, 4.9942e-03, 2.3133e-03]],
[[ 6.3604e-03, 6.0603e-04, 4.6913e-03],
[ 1.8093e-02, 5.0797e-03, -2.1611e-03],
[ 2.0538e-02, 8.5691e-03, -4.9416e-04]],
...,
[[-1.4583e-02, -1.8952e-03, -9.9537e-03],
[ 3.2319e-03, 1.1480e-02, -2.0378e-04],
[-6.8825e-03, -2.2193e-02, -2.6134e-02]],
[[ 1.8739e-02, 5.0367e-03, 3.4698e-03],
[ 2.9795e-04, 1.4573e-02, 5.4955e-04],
[ 7.8781e-03, 1.0652e-02, 8.1486e-04]],
[[ 1.8247e-03, -4.9992e-03, 7.0484e-03],
[-2.2999e-02, -1.8280e-02, -8.2337e-03],
[ 1.1130e-02, 9.6509e-03, -9.9123e-03]]],
[[[ 9.9050e-03, 4.4067e-04, 5.1800e-03],
[-1.0217e-02, 6.4841e-03, -3.8916e-03],
[ 4.3049e-03, 4.3258e-03, 2.2527e-03]],
[[ 9.3521e-03, 6.7698e-03, 8.7522e-04],
[-4.9717e-03, -2.2795e-02, -7.7713e-03],
[ 1.3400e-02, -6.3761e-03, 8.1015e-03]],
[[-9.7663e-03, -9.9591e-03, 1.9212e-03],
[-2.8908e-03, -1.3595e-02, 3.3229e-03],
[ 1.0392e-02, 1.1982e-02, 1.5520e-02]],
...,
[[ 1.1608e-02, 2.0908e-03, 1.7663e-02],
[ 1.6227e-02, -4.2351e-03, 2.6666e-03],
[ 1.3186e-02, 5.4400e-03, 7.4682e-03]],
[[-8.6877e-03, -1.8332e-02, -2.1043e-02],
[-9.4582e-03, -7.6206e-03, -1.2001e-02],
[-1.9868e-03, -2.1778e-02, -1.0427e-02]],
[[ 8.3325e-03, 1.7789e-02, 2.0029e-02],
[ 1.1773e-02, 3.6510e-02, 2.0913e-02],
[-1.1721e-02, -3.7434e-04, -1.5423e-02]]],
[[[ 2.5556e-06, 1.2850e-03, 1.2534e-03],
[-5.3958e-03, 2.3388e-02, 8.0807e-03],
[-1.0583e-02, 1.1113e-02, -1.0231e-02]],
[[ 1.2460e-02, 3.9356e-03, 4.9127e-04],
[ 4.1070e-02, 5.5333e-03, -1.7081e-02],
[ 2.5684e-02, -5.8720e-05, -4.9908e-03]],
[[ 2.3689e-03, -9.2193e-03, -1.6447e-03],
[-8.1313e-03, -6.4708e-03, 1.3475e-02],
[ 3.4459e-03, 5.3489e-03, 2.1980e-02]],
...,
[[ 1.2527e-02, -1.6218e-02, -1.8189e-02],
[ 5.9836e-02, -9.9851e-03, -4.4260e-02],
[ 3.1816e-02, -7.3521e-03, -2.8641e-02]],
[[-5.4672e-03, 6.5547e-03, 1.8037e-02],
[-1.2493e-03, 1.8533e-03, 1.8728e-02],
[-1.0131e-02, 2.2487e-03, 1.1088e-02]],
[[-2.9215e-04, -1.0023e-02, -2.2871e-02],
[ 1.4144e-02, -1.2906e-03, -1.3304e-02],
[ 8.5294e-03, -6.6538e-03, 1.4985e-03]]]], device='cuda:0')), ('layer3.1.bn2.weight', tensor([ 0.1731, 0.2074, 0.1871, 0.1670, 0.1743, 0.1365, 0.2004,
0.1824, 0.2402, 0.1968, 0.1630, 0.4164, 0.1519, 0.1665,
0.2208, 0.1864, 0.1493, 0.1711, 0.1702, 0.2000, 0.2133,
0.2083, 0.1884, 0.2025, 0.1830, 0.1845, 0.2036, 0.1779,
0.3539, 0.2236, 0.1405, 0.2074, 0.2006, 0.2243, 0.2032,
0.1363, 0.1512, 0.1482, 0.2272, 0.2192, 0.1916, 0.2044,
0.2011, 0.2135, 0.2072, 0.2053, 0.2050, 0.2323, 0.1812,
0.1631, 0.1622, 0.1982, 0.1931, 0.1841, 0.2158, 0.4189,
0.1667, 0.1853, 0.1501, 0.1280, 0.1877, 0.1226, 0.1902,
0.1675, 0.2236, 0.2411, 0.1310, 0.1536, 0.1590, 0.1794,
0.1336, 0.1989, 0.2203, 0.1763, 0.2166, 0.2028, 0.1512,
0.1346, 0.1707, 0.1583, 0.1901, 0.1736, 0.1830, 0.1583,
0.3012, 0.2218, 0.1625, 0.2095, 0.1862, 0.1600, 0.1737,
0.2166, 0.2180, 0.1315, 0.1386, 0.1776, 0.2055, 0.2165,
0.1934, 0.1610, 0.1864, 0.2007, 0.1906, 0.2107, 0.2152,
0.2188, 0.1987, 0.1693, 0.2073, 0.1779, 0.1385, 0.1956,
0.2203, 0.2052, 0.1521, 0.1754, 0.1018, 0.1567, 0.1643,
0.2062, 0.1977, 0.1523, 0.1541, 0.1947, 0.1393, 0.2942,
0.1816, 0.2287, 0.1166, 0.2058, 0.2029, 0.1816, 0.1845,
0.1960, 0.1667, 0.1753, 0.1789, 0.3119, 0.1645, 0.1912,
0.1760, 0.1181, 0.1657, 0.1974, 0.1536, 0.1472, 0.1989,
0.1978, 0.1810, 0.1807, 0.1541, 0.1739, 0.1682, 0.1769,
0.1755, 0.1692, 0.1688, 0.1553, 0.1559, 0.1652, 0.2011,
0.1417, 0.2715, 0.2047, 0.1816, 0.2231, 0.1999, 0.2152,
0.1256, 0.1918, 0.1789, 0.1762, 0.1772, 0.1765, 0.2098,
0.1923, 0.1752, 0.1506, 0.1658, 0.1828, 0.2622, 0.1481,
0.1924, 0.2001, 0.1969, 0.1991, 0.1572, 0.2127, 0.1484,
0.1429, 0.1161, 0.1614, 0.2227, 0.1616, 0.2000, 0.1774,
0.1983, 0.1605, 0.1347, 0.1104, 0.2107, 0.1986, 0.2226,
0.2348, 0.1589, 0.1757, 0.1925, 0.1488, 0.1878, 0.1897,
0.1988, 0.1849, 0.1856, 0.2192, 0.2019, 0.2119, 0.2107,
0.1189, 0.2126, 0.1324, 0.1402, 0.2058, 0.1748, 0.1862,
0.1895, 0.1913, 0.1620, 0.2373, 0.2028, 0.1856, 0.2267,
0.1578, 0.1179, 0.1834, 0.1901, 0.1920, 0.1785, 0.2026,
0.1830, 0.1807, 0.1281, 0.1472, 0.1712, 0.1617, 0.1966,
0.1334, 0.1619, 0.1734, 0.2112, 0.1783, 0.2074, 0.2375,
0.1459, 0.2741, 0.1850, 0.2070], device='cuda:0')), ('layer3.1.bn2.bias', tensor([-0.0682, -0.0851, -0.0391, -0.0351, -0.0896, 0.1120, -0.1100,
-0.0800, -0.2344, -0.0984, -0.0308, -0.3539, -0.0335, -0.0437,
-0.0521, -0.0073, -0.0157, -0.0517, -0.0711, -0.0671, -0.0659,
-0.0882, -0.0913, -0.0722, -0.0784, 0.0092, -0.0671, -0.0151,
-0.3271, -0.0666, 0.0887, -0.0824, -0.0853, -0.0969, -0.0878,
0.0835, -0.0145, -0.0468, -0.0697, -0.1085, -0.0448, -0.0667,
-0.0889, -0.0864, -0.0976, -0.1196, -0.0993, -0.1718, -0.1018,
0.0340, 0.0450, -0.0912, -0.0546, -0.0343, -0.0963, -0.3843,
-0.1074, -0.1241, 0.0003, -0.0112, -0.0981, 0.0642, 0.0102,
-0.0269, -0.1398, -0.0941, 0.0107, -0.0116, 0.0233, -0.0522,
0.0125, -0.0986, -0.0823, -0.0231, -0.1280, -0.0397, -0.0063,
0.0787, -0.0260, 0.0622, -0.0408, -0.0391, -0.0817, 0.0169,
-0.1824, -0.0140, 0.0718, -0.1180, -0.0929, 0.0562, -0.0665,
-0.1822, -0.1039, 0.0412, 0.0474, -0.0959, -0.1025, -0.1063,
-0.0606, -0.0087, -0.0768, -0.0650, -0.0905, -0.0862, -0.1085,
-0.1113, -0.0962, -0.0326, -0.1589, -0.0490, 0.0086, -0.0740,
-0.1155, -0.0948, 0.0280, -0.0645, 0.1358, 0.1967, 0.0292,
-0.0370, -0.0452, -0.0463, -0.0176, -0.0805, 0.0257, -0.2905,
-0.0805, -0.1202, 0.0980, -0.1287, -0.1120, -0.0723, -0.1051,
-0.1412, -0.0519, -0.0583, -0.0315, -0.2446, -0.0379, -0.0453,
0.0040, 0.1514, -0.0498, -0.0638, 0.0038, 0.0538, -0.0915,
-0.0760, 0.0302, -0.0889, -0.0263, -0.0875, 0.0158, -0.0589,
-0.0357, -0.0569, 0.0634, 0.0105, -0.0070, 0.0627, -0.0758,
0.0863, -0.2153, -0.0810, -0.0738, -0.1313, -0.0570, -0.1886,
0.0618, -0.0812, -0.0471, -0.0338, -0.0388, -0.0150, -0.0710,
-0.1072, -0.1244, 0.0584, -0.0344, -0.0279, -0.0645, -0.0091,
-0.0496, -0.0857, -0.1132, -0.1535, -0.0086, -0.0866, -0.0175,
0.0114, 0.1744, -0.0258, -0.1125, -0.0245, -0.0331, -0.0503,
-0.1195, 0.0278, 0.0685, 0.1036, -0.0832, -0.1362, -0.1125,
-0.0856, -0.0136, -0.0638, -0.0670, 0.0159, -0.1076, -0.1329,
-0.0898, -0.0535, -0.0888, -0.1127, -0.1038, -0.0894, -0.0815,
0.0355, -0.0972, 0.0592, 0.0117, -0.0993, -0.0567, 0.0058,
-0.0914, -0.0783, 0.0089, -0.1392, -0.0782, -0.1044, -0.0695,
-0.0227, 0.0968, -0.0742, -0.0867, -0.0921, -0.0921, -0.0667,
-0.0957, -0.0875, 0.0540, -0.0079, -0.0345, 0.0074, -0.1009,
0.0737, -0.0077, -0.0230, -0.0790, -0.0560, -0.0897, -0.1555,
0.0294, -0.2282, -0.0963, -0.1058], device='cuda:0')), ('layer3.1.bn2.running_mean', tensor([ 0.1293, -0.0940, 0.0104, -0.0428, 0.0081, 0.1639, -0.0357,
0.5185, -0.0765, -0.0626, 0.0201, -0.0626, 0.0606, 0.0326,
-0.0852, -0.0030, 0.1104, -0.0266, 0.0444, -0.2011, 0.0029,
-0.0402, -0.0854, -0.0621, -0.1063, 0.1064, -0.1454, -0.0522,
0.1140, -0.7557, 0.1762, 0.0417, 0.0331, -0.0640, 0.2039,
0.2325, 0.0501, -0.1193, -0.2955, -0.0762, -0.0580, -0.0583,
-0.0174, -0.1112, -0.0678, 0.0065, -0.0494, -0.0475, -0.0932,
0.0915, 0.1216, -0.0952, 0.0663, -0.1623, -0.0961, -0.3499,
-0.0185, -0.0147, 0.0640, -0.0064, -0.0606, -0.0164, 0.0512,
0.0149, -0.1429, -0.1316, 0.0583, 0.0754, 0.0332, -0.0342,
-0.0148, -0.0458, -0.1083, -0.0431, -0.0343, 0.0118, 0.1081,
0.1879, 0.0005, 0.2422, -0.0212, 0.1496, -0.0332, -0.2118,
-0.1503, -0.0314, 0.2240, -0.0578, -0.1089, -0.0055, -0.0467,
0.1456, -0.0336, 0.0712, -0.0246, 0.1026, -0.0608, -0.1612,
0.0181, -0.0160, 0.0787, 0.0009, -0.0433, -0.0272, -0.0397,
-0.0717, -0.0750, -0.1143, 0.1782, 0.0569, -0.0159, -0.0610,
-0.2420, -0.0927, 0.0504, 0.0083, 0.5746, -0.3408, 0.0498,
0.0386, -0.0296, -0.0427, 0.0494, -0.1592, -0.1777, -0.1993,
-0.0634, -0.0520, -0.1304, -0.1120, -0.1009, -0.0068, -0.0143,
-0.0528, -0.1016, -0.0897, 0.0213, -0.3422, 0.1069, -0.2116,
0.0506, 0.4207, 0.0417, -0.0569, 0.0931, 0.0818, -0.1563,
-0.2065, -0.0910, -0.1511, 0.0320, 0.0298, 0.0240, 0.0479,
0.1176, -0.0058, -0.1015, -0.1814, -0.1129, -0.1753, -0.0441,
-0.1445, -0.0715, -0.0346, -0.1082, -0.1920, -0.0657, -0.0328,
0.0308, 0.1938, -0.0684, 0.0056, -0.0068, 0.0112, -0.1918,
0.0007, -0.0497, -0.1370, 0.0207, 0.0242, -0.8942, -0.1884,
-0.0657, -0.0900, -0.0540, 0.0985, -0.1212, -0.0719, -0.0283,
-0.1557, 0.1516, -0.0277, -0.0857, -0.0308, -0.0502, -0.0582,
-0.0316, 0.0065, -0.0982, 0.0274, -0.0778, -0.1195, -0.0643,
-0.0443, 0.0154, 0.0014, -0.0074, 0.0331, 0.0254, -0.0924,
-0.0126, 0.0243, -0.0126, 0.0860, -0.0498, -0.0537, -0.0109,
-0.0626, -0.0718, 0.0978, -0.0630, -0.0428, -0.0367, -0.1610,
-0.0299, -0.0454, 0.0123, -0.0449, -0.2119, 0.1048, -0.1404,
-0.0112, 0.1883, 0.0048, -0.2282, -0.0463, -0.0702, 0.0273,
-0.0640, -0.0810, -0.2001, 0.0158, 0.0105, -0.0254, 0.0394,
-0.0678, -0.0759, 0.0631, -0.1309, 0.0567, -0.0032, -0.1765,
0.0250, -0.2393, 0.0106, 0.0494], device='cuda:0')), ('layer3.1.bn2.running_var', tensor([ 0.0157, 0.0204, 0.0263, 0.0121, 0.0134, 0.0405, 0.0202,
0.0148, 0.0227, 0.0173, 0.0267, 0.0434, 0.0181, 0.0214,
0.0611, 0.0321, 0.0161, 0.0193, 0.0187, 0.0194, 0.0380,
0.0401, 0.0138, 0.0208, 0.0149, 0.0323, 0.0201, 0.0349,
0.1201, 0.0563, 0.0380, 0.0241, 0.0172, 0.0279, 0.0211,
0.0317, 0.0085, 0.0125, 0.0266, 0.0165, 0.0184, 0.0303,
0.0204, 0.0405, 0.0194, 0.0184, 0.0187, 0.0138, 0.0104,
0.0246, 0.0433, 0.0218, 0.0184, 0.0137, 0.0345, 0.0597,
0.0092, 0.0146, 0.0214, 0.0113, 0.0153, 0.0141, 0.0256,
0.0159, 0.0130, 0.0398, 0.0106, 0.0313, 0.0324, 0.0151,
0.0144, 0.0186, 0.0177, 0.0401, 0.0237, 0.0296, 0.0213,
0.0307, 0.0237, 0.0339, 0.0226, 0.0249, 0.0107, 0.0183,
0.0264, 0.0504, 0.0448, 0.0165, 0.0160, 0.0375, 0.0112,
0.0144, 0.0401, 0.0128, 0.0223, 0.0169, 0.0185, 0.0194,
0.0162, 0.0188, 0.0150, 0.0173, 0.0159, 0.0305, 0.0201,
0.0355, 0.0215, 0.0221, 0.0168, 0.0135, 0.0163, 0.0154,
0.0168, 0.0237, 0.0217, 0.0179, 0.0167, 0.0376, 0.0416,
0.0266, 0.0150, 0.0090, 0.0213, 0.0158, 0.0149, 0.0167,
0.0161, 0.0345, 0.0140, 0.0218, 0.0135, 0.0178, 0.0164,
0.0140, 0.0130, 0.0137, 0.0189, 0.0302, 0.0283, 0.0145,
0.0228, 0.0180, 0.0221, 0.0248, 0.0164, 0.0251, 0.0140,
0.0244, 0.0266, 0.0147, 0.0121, 0.0128, 0.0197, 0.0260,
0.0242, 0.0140, 0.0303, 0.0146, 0.0159, 0.0297, 0.0265,
0.0325, 0.0172, 0.0355, 0.0216, 0.0157, 0.0289, 0.0136,
0.0249, 0.0282, 0.0176, 0.0341, 0.0224, 0.0210, 0.0202,
0.0225, 0.0100, 0.0257, 0.0219, 0.0163, 0.0748, 0.0137,
0.0140, 0.0325, 0.0171, 0.0183, 0.0157, 0.0293, 0.0118,
0.0149, 0.0145, 0.0178, 0.0198, 0.0106, 0.0272, 0.0154,
0.0135, 0.0209, 0.0213, 0.0140, 0.0252, 0.0203, 0.0216,
0.0361, 0.0150, 0.0163, 0.0224, 0.0101, 0.0255, 0.0150,
0.0218, 0.0266, 0.0225, 0.0172, 0.0200, 0.0235, 0.0263,
0.0116, 0.0187, 0.0224, 0.0212, 0.0134, 0.0110, 0.0307,
0.0217, 0.0328, 0.0187, 0.0216, 0.0175, 0.0156, 0.0485,
0.0157, 0.0281, 0.0286, 0.0130, 0.0145, 0.0130, 0.0260,
0.0200, 0.0137, 0.0217, 0.0162, 0.0227, 0.0173, 0.0178,
0.0246, 0.0135, 0.0214, 0.0157, 0.0126, 0.0160, 0.0140,
0.0239, 0.0255, 0.0126, 0.0191], device='cuda:0')), ('layer3.1.conv3.weight', tensor([[[[-6.3724e-03]],
[[ 2.3534e-02]],
[[ 2.4592e-03]],
...,
[[ 4.1078e-03]],
[[ 2.4874e-03]],
[[-1.0990e-02]]],
[[[-8.3257e-03]],
[[ 1.2592e-02]],
[[-8.2291e-03]],
...,
[[ 1.0739e-02]],
[[-7.6186e-03]],
[[ 1.3689e-02]]],
[[[-1.0180e-02]],
[[-9.5879e-03]],
[[-5.3071e-04]],
...,
[[ 1.5783e-02]],
[[ 1.0041e-02]],
[[ 1.8456e-03]]],
...,
[[[ 3.2607e-02]],
[[-9.3498e-03]],
[[ 2.1410e-02]],
...,
[[ 4.0840e-02]],
[[-1.1005e-02]],
[[ 4.2452e-02]]],
[[[ 1.2486e-02]],
[[ 6.3653e-03]],
[[-1.2754e-02]],
...,
[[-1.4025e-02]],
[[-7.0006e-03]],
[[-1.3153e-02]]],
[[[-1.4763e-02]],
[[ 2.1255e-02]],
[[ 5.7463e-03]],
...,
[[-6.0617e-03]],
[[-6.3444e-03]],
[[-8.2349e-03]]]], device='cuda:0')), ('layer3.1.bn3.weight', tensor([ 4.8032e-02, 8.2487e-02, 1.1153e-01, ..., 1.3482e-01,
1.0690e-01, 1.0801e-01], device='cuda:0')), ('layer3.1.bn3.bias', tensor([-0.0647, -0.0260, -0.0057, ..., -0.0232, -0.0627, -0.0947], device='cuda:0')), ('layer3.1.bn3.running_mean', tensor([-3.9994e-02, -1.7666e-04, 1.1486e-02, ..., 1.4528e-02,
1.4916e-02, -3.5621e-02], device='cuda:0')), ('layer3.1.bn3.running_var', tensor([ 7.8426e-04, 8.5304e-04, 1.3059e-03, ..., 1.6498e-03,
1.7968e-03, 1.8496e-03], device='cuda:0')), ('layer3.2.conv1.weight', tensor([[[[ 2.8827e-03]],
[[-2.8368e-02]],
[[-2.5213e-03]],
...,
[[-4.1888e-02]],
[[-2.7244e-02]],
[[-1.0362e-02]]],
[[[-1.2423e-02]],
[[-3.0724e-03]],
[[ 1.1820e-02]],
...,
[[ 4.6387e-02]],
[[-3.1436e-03]],
[[ 2.2673e-02]]],
[[[ 3.5264e-03]],
[[-2.7772e-02]],
[[ 1.9115e-02]],
...,
[[-6.3041e-03]],
[[-3.1398e-02]],
[[ 1.3827e-02]]],
...,
[[[ 5.8742e-03]],
[[ 5.8682e-03]],
[[ 3.8751e-03]],
...,
[[-1.2832e-02]],
[[-1.6594e-02]],
[[ 1.1244e-02]]],
[[[ 1.5227e-02]],
[[-9.1758e-03]],
[[-8.0038e-03]],
...,
[[-1.0496e-02]],
[[-1.5176e-02]],
[[ 2.1703e-02]]],
[[[ 1.6154e-02]],
[[ 1.6533e-02]],
[[ 6.5476e-03]],
...,
[[-1.4464e-02]],
[[ 1.0980e-03]],
[[-1.0425e-02]]]], device='cuda:0')), ('layer3.2.bn1.weight', tensor([ 0.2235, 0.1873, 0.1171, 0.1336, 0.1773, 0.1729, 0.1879,
0.1848, 0.1492, 0.1673, 0.1416, 0.1602, 0.1017, 0.0980,
0.1936, 0.1305, 0.1814, 0.1877, 0.1290, 0.1670, 0.1490,
0.1928, 0.1335, 0.1712, 0.2340, 0.2333, 0.1540, 0.1447,
0.1428, 0.2174, 0.1549, 0.1337, 0.1050, 0.1420, 0.1685,
0.1133, 0.1701, 0.1309, 0.1737, 0.1704, 0.1224, 0.1849,
0.1040, 0.1580, 0.1581, 0.1564, 0.1767, 0.1913, 0.1188,
0.1222, 0.1635, 0.1957, 0.1553, 0.2126, 0.1243, 0.1475,
0.1571, 0.1037, 0.1739, 0.1467, 0.1578, 0.1570, 0.1661,
0.1471, 0.1935, 0.1229, 0.1083, 0.1505, 0.1414, 0.1675,
0.1396, 0.2291, 0.1739, 0.1506, 0.1584, 0.1762, 0.1559,
0.1363, 0.1720, 0.1711, 0.1619, 0.1372, 0.1843, 0.1320,
0.1509, 0.1483, 0.1984, 0.1078, 0.1616, 0.2366, 0.1402,
0.1534, 0.1739, 0.1415, 0.1707, 0.1558, 0.1246, 0.1815,
0.1427, 0.2065, 0.1840, 0.1461, 0.1611, 0.1643, 0.1481,
0.2103, 0.1430, 0.1069, 0.1375, 0.1731, 0.1506, 0.1766,
0.1287, 0.0985, 0.1829, 0.1552, 0.1715, 0.0993, 0.1151,
0.1370, 0.1645, 0.1582, 0.1257, 0.1490, 0.1467, 0.1230,
0.1268, 0.1470, 0.1401, 0.1523, 0.1669, 0.1628, 0.1419,
0.1525, 0.1438, 0.1671, 0.1612, 0.1437, 0.1443, 0.1590,
0.1501, 0.1948, 0.1376, 0.1607, 0.1591, 0.1243, 0.1584,
0.1740, 0.1976, 0.1319, 0.1418, 0.1325, 0.1438, 0.1928,
0.1405, 0.1653, 0.1511, 0.1458, 0.1797, 0.1337, 0.1317,
0.1902, 0.1464, 0.1427, 0.1532, 0.1824, 0.1675, 0.1986,
0.1489, 0.1340, 0.1183, 0.1376, 0.1944, 0.1688, 0.1288,
0.1371, 0.1695, 0.1649, 0.1248, 0.1500, 0.1819, 0.1421,
0.1907, 0.1399, 0.1824, 0.1318, 0.1926, 0.1596, 0.1611,
0.1655, 0.1560, 0.1445, 0.1513, 0.1360, 0.1715, 0.1498,
0.1814, 0.1285, 0.1634, 0.2138, 0.1419, 0.1172, 0.1420,
0.1975, 0.1398, 0.1707, 0.1561, 0.1520, 0.1897, 0.1404,
0.2061, 0.1286, 0.1501, 0.1298, 0.1451, 0.1348, 0.1901,
0.1154, 0.1924, 0.1203, 0.1630, 0.1382, 0.1270, 0.1348,
0.1255, 0.1802, 0.1576, 0.1258, 0.1703, 0.1120, 0.1179,
0.1749, 0.1213, 0.1250, 0.1880, 0.1126, 0.1604, 0.1708,
0.1650, 0.1671, 0.1411, 0.1908, 0.1489, 0.1521, 0.2002,
0.1725, 0.1181, 0.1745, 0.1503, 0.1096, 0.1502, 0.1724,
0.1517, 0.1597, 0.1411, 0.1436], device='cuda:0')), ('layer3.2.bn1.bias', tensor([-0.1153, -0.1992, 0.0628, 0.0422, -0.0964, -0.1043, -0.0946,
-0.1055, -0.0161, -0.0874, -0.0508, -0.0519, 0.1040, 0.0593,
-0.0399, 0.0009, -0.0857, -0.1124, 0.0024, -0.0689, -0.0599,
-0.1073, 0.0200, -0.0628, -0.1832, -0.1608, -0.0777, -0.0135,
-0.0229, -0.1460, -0.1282, -0.0234, 0.0851, -0.0166, -0.0403,
0.0237, -0.0470, 0.0321, -0.1158, -0.0258, 0.0330, -0.1253,
0.0473, -0.0672, -0.0544, 0.0179, -0.0576, -0.1480, 0.0362,
-0.0002, -0.0598, -0.0935, -0.1279, -0.1138, 0.0243, 0.0085,
-0.0976, 0.1186, -0.0898, -0.0741, -0.0487, -0.0462, -0.0780,
-0.0190, -0.0050, 0.0345, 0.0936, -0.0416, -0.0457, -0.1221,
-0.0061, -0.2316, -0.0468, -0.0584, -0.0322, -0.1026, -0.0254,
-0.0301, -0.0734, -0.0109, -0.0446, 0.0218, -0.0049, 0.0203,
-0.0990, 0.0190, -0.1278, 0.0519, -0.0292, -0.0879, 0.0071,
-0.0577, -0.0466, -0.0300, -0.1127, -0.0791, -0.0110, -0.1309,
-0.0399, -0.1383, -0.1191, -0.0569, -0.0807, -0.0822, 0.0002,
-0.2088, 0.0078, 0.0311, -0.0306, -0.0592, -0.0380, -0.1168,
0.0270, 0.0711, -0.0910, -0.0429, -0.0818, 0.1021, 0.0863,
-0.0272, -0.0480, -0.0459, -0.0025, -0.0241, -0.0229, -0.0039,
0.0128, -0.0811, -0.0465, -0.0577, -0.0664, -0.0260, 0.0169,
-0.0575, -0.0408, -0.0583, -0.0747, -0.0098, -0.0481, -0.0953,
-0.0299, -0.0947, -0.0090, -0.0821, -0.0753, 0.0148, -0.0669,
-0.0864, -0.1115, 0.0252, -0.0163, -0.0107, -0.0034, -0.1146,
-0.0008, -0.1254, -0.0722, -0.0411, -0.0878, -0.0354, 0.0252,
-0.1397, -0.0852, -0.0475, -0.1516, -0.1362, -0.0456, -0.1530,
-0.0146, 0.0129, 0.0501, -0.0238, -0.1141, -0.0091, 0.0130,
-0.0172, -0.0405, -0.0968, 0.0388, -0.0249, -0.0939, -0.0435,
-0.0896, -0.0161, -0.0463, -0.0165, -0.1177, -0.0453, -0.0835,
-0.0824, -0.0726, -0.0417, -0.0990, -0.0350, -0.0870, -0.0277,
-0.1810, 0.0337, 0.0543, -0.1367, -0.0697, 0.0301, -0.0135,
-0.1152, -0.0763, -0.0659, -0.0534, -0.0412, -0.1344, -0.0177,
-0.1512, -0.0178, -0.1142, -0.0093, -0.0622, -0.0654, -0.1472,
0.0123, -0.1172, 0.0396, -0.0550, -0.0423, 0.0328, -0.0191,
0.1064, -0.0650, -0.0401, -0.0076, -0.0657, 0.0405, 0.0278,
-0.1149, 0.0298, -0.0368, -0.0929, 0.0423, -0.0758, -0.1033,
-0.0653, -0.0877, -0.0171, -0.1000, -0.0304, -0.0560, -0.1225,
-0.0064, 0.0038, -0.0925, -0.0310, 0.0688, -0.0719, -0.0698,
0.0347, -0.0650, -0.0171, -0.0507], device='cuda:0')), ('layer3.2.bn1.running_mean', tensor([-0.1376, 0.0372, 0.0041, -0.0050, -0.1340, -0.0177, -0.0186,
-0.0787, -0.0361, -0.1014, 0.0187, -0.0999, -0.0698, -0.1767,
-0.1084, -0.0709, -0.1230, -0.0200, -0.0735, -0.1864, -0.0018,
-0.1371, -0.0857, -0.0616, -0.1970, -0.0220, -0.1086, -0.0638,
-0.0776, -0.1190, 0.0128, 0.0046, -0.0553, -0.0978, -0.0448,
-0.0438, -0.0389, -0.0531, 0.0912, 0.0435, -0.0814, -0.1097,
0.0064, -0.0039, -0.0862, -0.1205, -0.0125, 0.0338, -0.0938,
-0.0246, -0.0423, -0.0841, -0.0253, -0.0554, -0.0012, 0.0254,
-0.0232, -0.0235, -0.0164, 0.0103, 0.0739, 0.0301, 0.0039,
0.0269, 0.0420, -0.0479, -0.1608, -0.0546, -0.0048, -0.0638,
-0.0215, -0.0655, 0.0241, -0.0639, -0.0921, -0.0539, -0.0863,
-0.0670, -0.0098, -0.0727, -0.1201, 0.0858, -0.1827, -0.0331,
-0.0123, -0.1129, -0.0804, 0.0136, -0.0761, 0.0151, -0.0947,
0.0086, -0.0243, -0.0709, -0.0700, 0.0108, -0.0981, -0.0539,
0.0344, -0.1476, -0.0325, -0.0552, 0.0501, -0.0355, 0.0144,
-0.0103, -0.0787, -0.1924, -0.0820, 0.0197, -0.1441, -0.1140,
-0.0327, 0.0391, -0.0309, -0.0053, -0.0204, -0.1595, -0.0437,
-0.0633, -0.0602, 0.0254, -0.1934, 0.0042, -0.1514, -0.1380,
-0.1992, -0.0264, 0.0026, -0.0571, 0.0363, -0.0692, 0.0034,
0.0474, 0.0050, 0.1018, -0.1211, -0.1177, -0.0895, -0.1229,
-0.0789, -0.0875, 0.0058, 0.0436, -0.0035, -0.1242, -0.0133,
0.0612, 0.0485, 0.0242, -0.1446, -0.0213, 0.0663, -0.0851,
-0.0888, 0.0330, -0.0039, -0.0173, -0.0546, -0.0462, 0.0119,
-0.0846, -0.0579, -0.0382, 0.0426, -0.0268, -0.0547, -0.0209,
-0.0294, -0.2139, 0.0216, 0.0006, -0.1584, 0.0314, -0.0026,
-0.0233, -0.1612, 0.0615, 0.0418, -0.1625, -0.1001, -0.0825,
0.0017, -0.0811, -0.0219, -0.0065, -0.0845, -0.0352, -0.0829,
-0.0884, -0.1230, -0.0716, 0.0538, -0.1006, -0.0945, -0.0598,
0.0737, -0.0562, 0.0047, 0.0077, -0.0299, -0.1322, -0.1367,
-0.0496, 0.0414, -0.0708, -0.0143, -0.1074, -0.1350, -0.0522,
-0.1582, -0.2141, 0.0272, -0.0312, -0.0193, -0.0624, -0.0004,
-0.0857, -0.1005, -0.1001, -0.0476, -0.0184, 0.0161, 0.0064,
0.0331, -0.0432, 0.0229, -0.0112, -0.0002, -0.0813, 0.0009,
-0.0949, -0.0835, -0.0438, 0.2036, -0.1581, -0.0258, -0.0783,
-0.1881, -0.0919, -0.1272, -0.1325, -0.0420, -0.0806, -0.0945,
-0.1294, -0.1484, -0.0243, -0.1068, -0.0477, -0.0449, -0.1772,
-0.1617, -0.0607, -0.0891, -0.0496], device='cuda:0')), ('layer3.2.bn1.running_var', tensor(1.00000e-02 *
[ 2.1460, 0.8076, 1.7251, 2.8212, 1.8435, 0.8798, 1.8702,
2.1261, 1.5667, 1.3873, 1.5700, 1.2852, 2.3662, 1.2656,
3.0820, 1.8936, 2.0139, 1.3481, 1.5878, 1.2191, 0.9833,
1.6628, 1.6912, 0.9979, 1.9055, 1.6320, 1.8952, 1.5827,
2.3570, 1.7628, 0.9237, 1.7783, 1.5303, 1.6382, 2.2555,
1.6990, 2.2752, 2.1375, 1.3684, 2.0146, 2.9448, 3.2426,
2.5991, 1.0814, 1.8170, 2.5719, 1.8144, 0.8012, 1.7499,
1.3977, 1.8768, 1.4086, 1.1558, 3.0632, 2.2223, 2.5223,
0.9318, 1.7581, 2.2350, 1.5356, 2.3987, 2.1265, 1.2320,
2.3234, 2.5962, 2.3019, 1.9131, 1.9615, 1.6462, 1.6345,
2.7716, 1.9753, 2.3134, 1.4048, 1.8070, 1.0592, 1.5110,
2.2447, 2.5455, 2.1518, 1.9668, 2.5028, 2.8221, 1.6488,
1.0722, 2.9769, 1.4412, 3.0675, 2.3319, 2.9616, 2.1855,
1.5858, 2.2188, 1.7830, 1.0585, 1.0777, 1.1600, 1.3304,
1.8512, 1.9857, 1.4382, 1.7218, 1.4866, 1.7250, 2.8348,
3.2091, 2.4563, 1.5789, 1.5517, 2.1502, 2.0227, 1.4481,
1.8272, 1.4387, 1.8601, 1.2382, 1.7329, 1.7489, 3.2982,
1.9232, 2.4790, 2.1416, 1.5788, 2.2256, 1.8140, 1.6912,
1.7140, 1.7104, 1.1494, 1.3892, 1.2228, 2.4342, 2.4326,
1.5668, 0.9960, 1.6552, 1.3820, 1.6649, 1.0380, 2.4498,
1.6481, 1.6069, 1.5982, 1.3331, 2.0661, 1.2159, 1.0860,
1.5903, 1.2439, 2.4000, 2.1058, 2.2167, 1.2983, 1.4746,
2.1243, 1.7242, 1.9167, 1.4606, 1.7823, 1.5214, 2.1811,
1.2167, 0.9594, 1.3280, 1.0562, 1.7846, 2.0229, 1.8535,
2.1851, 2.0973, 2.1332, 1.3176, 1.4566, 2.8554, 2.1368,
1.7114, 1.2484, 1.4232, 1.6476, 1.7091, 1.4672, 1.0881,
1.9891, 1.6482, 2.7423, 1.4256, 2.2208, 1.5721, 1.0289,
1.2228, 1.6446, 1.6236, 1.3420, 1.4004, 1.4423, 1.6302,
1.0867, 1.1959, 4.7939, 2.1998, 0.8402, 1.2541, 1.5859,
1.3652, 1.2641, 2.0189, 1.1261, 1.8978, 1.0516, 1.4707,
2.6648, 1.2551, 0.9449, 1.3292, 1.3998, 1.0285, 1.8665,
1.3425, 0.9911, 1.9442, 2.2240, 1.3033, 2.3173, 1.9223,
3.0440, 2.2724, 2.0288, 1.8278, 1.8481, 1.3140, 2.1036,
1.3756, 1.9086, 1.2901, 2.6710, 2.2531, 1.4336, 2.1847,
1.7471, 2.2171, 1.4617, 1.9936, 1.6408, 1.8122, 2.1050,
2.5642, 1.5179, 1.8130, 2.1420, 1.8105, 1.4313, 1.8507,
3.1338, 1.5972, 1.4669, 1.4694], device='cuda:0')), ('layer3.2.conv2.weight', tensor([[[[-3.5802e-03, -1.1028e-02, 1.3011e-03],
[ 1.1572e-02, 2.0107e-02, 1.2585e-02],
[ 1.8766e-02, 2.7120e-02, 2.3125e-02]],
[[ 8.0946e-03, 1.2307e-03, -5.1088e-03],
[ 6.1331e-03, 3.0108e-02, 8.6952e-03],
[-4.5609e-03, 8.1927e-03, -4.2226e-03]],
[[ 1.9080e-02, 1.8056e-02, 9.1025e-03],
[-2.0389e-02, -3.6315e-02, -1.7799e-02],
[ 1.5768e-02, 3.6663e-02, 1.7026e-02]],
...,
[[-2.1304e-02, -1.9290e-02, -1.3309e-02],
[-1.5200e-02, -9.1943e-03, 9.8947e-03],
[ 5.3930e-03, 4.1778e-02, 1.4718e-02]],
[[ 6.7867e-03, 1.7099e-02, 2.1589e-02],
[-5.3290e-03, -1.9471e-03, 1.4815e-02],
[-8.7725e-03, -1.3762e-02, 1.2047e-02]],
[[ 5.6504e-02, 2.7372e-02, -4.0524e-02],
[ 8.3591e-02, 8.5810e-03, -7.8745e-02],
[ 6.5760e-02, -1.1203e-03, -5.9587e-02]]],
[[[ 1.6778e-03, 2.1333e-03, -8.0604e-03],
[ 3.3667e-03, 3.3859e-02, 2.6999e-03],
[-4.1019e-03, 1.7711e-03, -7.0186e-04]],
[[-3.6651e-03, 5.4135e-03, 1.2987e-02],
[-2.2535e-02, -4.4939e-03, -9.9636e-03],
[-1.3400e-02, -1.9440e-02, -1.1785e-02]],
[[-2.0187e-03, 4.0392e-03, 8.7115e-03],
[-3.7967e-03, -1.9434e-03, 8.2896e-03],
[ 2.1840e-03, 7.2859e-03, 8.2192e-03]],
...,
[[-6.5260e-03, -1.0520e-02, 5.6257e-05],
[-8.5037e-03, -2.9023e-03, -2.1524e-03],
[-1.8586e-02, -1.3367e-02, -1.1288e-02]],
[[ 1.4722e-02, 1.2381e-02, 1.4357e-02],
[ 4.3434e-02, 2.2864e-02, 3.7622e-02],
[ 2.6757e-02, 2.9881e-02, 1.3510e-02]],
[[ 1.3036e-02, -3.8851e-04, -8.0173e-03],
[ 6.8872e-03, 1.0882e-04, -1.6729e-02],
[ 1.5520e-02, -8.3424e-05, -1.7248e-02]]],
[[[ 9.3224e-03, -8.5108e-03, -8.3709e-03],
[ 8.5115e-03, -1.0038e-02, 8.2710e-03],
[ 2.2448e-03, -8.4826e-04, 1.6983e-02]],
[[-1.1610e-02, -1.7194e-02, -1.3838e-02],
[-7.7607e-03, -6.2310e-03, -8.2235e-03],
[ 2.7842e-03, 2.8286e-03, -7.1709e-03]],
[[-1.3140e-03, -9.6142e-03, -1.5964e-03],
[ 5.9721e-04, -3.8326e-03, 2.8747e-03],
[-1.6226e-02, -1.0415e-02, -4.6902e-03]],
...,
[[ 1.7541e-02, 2.9383e-02, -1.8580e-03],
[-3.8670e-03, 6.0731e-03, -2.6307e-03],
[-3.4498e-02, -1.6476e-02, 1.0094e-02]],
[[-5.6360e-03, -5.3894e-03, -4.8574e-03],
[-1.0288e-02, -3.4400e-03, -4.3404e-03],
[ 1.7583e-03, 6.7847e-04, 1.0746e-02]],
[[-6.1896e-04, -2.7444e-03, -8.4192e-03],
[-9.1858e-03, 1.3946e-03, 7.9272e-03],
[-1.8049e-03, 2.5581e-03, 1.5893e-02]]],
...,
[[[ 1.2330e-02, -1.7113e-02, 4.6529e-03],
[ 1.1574e-02, -2.4131e-02, 1.3006e-02],
[ 1.3733e-02, -5.7683e-03, 8.1247e-03]],
[[ 2.5049e-03, -1.0780e-03, -3.9002e-03],
[ 1.6765e-02, 9.8542e-03, 1.4043e-02],
[-1.1201e-03, -6.0599e-03, -3.4802e-03]],
[[ 8.0176e-03, -1.0833e-02, 2.9448e-03],
[ 7.0284e-03, -1.6885e-02, -6.4530e-03],
[-1.5313e-03, -1.2809e-02, -1.7555e-02]],
...,
[[ 7.2650e-04, -3.4599e-02, 3.6178e-03],
[-1.1468e-03, -1.6426e-02, 8.2671e-03],
[ 2.9852e-03, -1.2011e-02, 2.7804e-03]],
[[ 1.3518e-02, -3.6790e-02, -2.8406e-03],
[-3.6350e-03, -3.7368e-02, -9.5802e-03],
[ 1.3845e-02, -2.0064e-02, 1.7846e-03]],
[[-2.9108e-03, 3.5288e-03, 1.1068e-02],
[-9.1812e-03, 2.1687e-04, 1.5875e-02],
[ 4.1262e-03, -1.6199e-04, 6.8643e-04]]],
[[[ 2.3126e-02, 1.7599e-02, 1.8742e-02],
[ 2.9786e-02, 1.1471e-02, 7.7411e-03],
[ 1.9394e-03, 3.7610e-03, -1.0510e-02]],
[[-2.6477e-02, -1.8215e-02, 1.5537e-02],
[-3.9110e-02, -3.6499e-02, 2.6256e-02],
[-3.0353e-02, -4.6805e-02, -1.6427e-02]],
[[-7.5252e-03, 2.2827e-03, 3.6183e-03],
[-1.2927e-02, -9.0444e-03, 2.2144e-02],
[-6.5457e-03, -1.3059e-02, -3.4611e-03]],
...,
[[-1.4686e-02, -6.7829e-03, -5.8789e-03],
[-1.1703e-02, -3.4396e-03, 8.6091e-04],
[-9.0684e-03, -4.9701e-03, 2.1739e-02]],
[[ 3.5613e-03, -5.1085e-03, -2.2686e-02],
[ 5.7618e-03, -5.8313e-04, -1.1181e-02],
[-7.6683e-03, -1.2790e-02, -8.8351e-03]],
[[-8.7081e-03, -2.4242e-03, 1.7277e-02],
[-6.8912e-03, -8.9876e-03, 7.6058e-03],
[-4.3620e-03, -1.2196e-02, -1.0771e-02]]],
[[[-1.1395e-02, -1.6485e-02, -8.1376e-03],
[-1.0096e-04, -3.0751e-02, 4.5764e-03],
[ 7.2135e-03, -1.9183e-02, -8.7631e-04]],
[[-2.5444e-02, -1.2167e-02, -1.5170e-02],
[-2.6833e-02, -1.3808e-02, -1.7042e-02],
[-2.0713e-02, -1.7758e-02, -1.7370e-02]],
[[-1.6505e-02, -8.0067e-03, 1.1924e-03],
[-1.7738e-02, -3.1269e-03, 1.2873e-02],
[-2.0834e-02, -1.8683e-03, 1.0609e-02]],
...,
[[ 4.2366e-03, 1.1488e-02, -7.5118e-03],
[-1.2653e-02, -4.1932e-03, -1.6243e-02],
[-1.8112e-02, -2.3787e-02, 4.6444e-03]],
[[-1.3521e-02, 4.4411e-03, -1.8358e-02],
[-3.1651e-02, 6.4925e-03, -2.0924e-02],
[-2.9287e-02, -6.6709e-04, -7.9085e-03]],
[[-4.8704e-03, -2.1171e-03, 4.6133e-03],
[-6.9341e-03, -2.1932e-02, 5.6086e-03],
[-2.1563e-05, -4.4543e-03, 2.2337e-02]]]], device='cuda:0')), ('layer3.2.bn2.weight', tensor([ 0.0969, 0.1814, 0.1938, 0.1642, 0.2110, 0.2123, 0.1423,
0.1863, 0.1374, 0.1904, 0.1703, 0.1753, 0.2232, 0.1309,
0.1904, 0.1800, 0.2098, 0.2059, 0.2137, 0.1609, 0.1975,
0.2097, 0.1341, 0.1360, 0.1813, 0.2388, 0.1784, 0.1816,
0.1364, 0.2038, 0.1830, 0.2292, 0.2131, 0.2544, 0.2114,
0.1966, 0.1961, 0.1324, 0.1402, 0.2292, 0.1947, 0.2015,
0.1747, 0.1883, 0.1947, 0.1927, 0.2368, 0.2083, 0.2113,
0.1313, 0.1306, 0.1675, 0.1935, 0.1699, 0.1958, 0.1847,
0.1984, 0.1885, 0.1834, 0.1234, 0.2094, 0.1791, 0.1495,
0.1383, 0.1268, 0.1357, 0.2262, 0.1566, 0.1690, 0.1925,
0.2146, 0.1678, 0.1544, 0.1702, 0.2020, 0.2086, 0.2057,
0.2180, 0.1834, 0.1976, 0.1290, 0.1304, 0.2164, 0.1939,
0.1878, 0.1262, 0.1837, 0.2008, 0.1922, 0.2035, 0.2134,
0.2244, 0.2234, 0.1850, 0.1372, 0.1804, 0.2006, 0.2246,
0.1644, 0.1731, 0.1765, 0.2073, 0.1688, 0.1802, 0.2063,
0.2126, 0.2001, 0.2032, 0.1740, 0.2060, 0.2358, 0.1634,
0.1929, 0.1728, 0.1168, 0.1076, 0.2326, 0.2094, 0.1260,
0.1913, 0.2373, 0.2136, 0.1605, 0.1765, 0.1873, 0.1220,
0.2000, 0.1367, 0.1592, 0.1939, 0.1984, 0.2398, 0.1705,
0.2116, 0.1427, 0.1169, 0.1843, 0.1933, 0.1586, 0.2246,
0.1880, 0.1705, 0.1947, 0.1136, 0.1762, 0.2038, 0.1890,
0.1363, 0.2078, 0.1812, 0.1731, 0.2160, 0.2013, 0.1872,
0.1884, 0.1926, 0.2031, 0.1792, 0.1819, 0.2186, 0.2463,
0.1658, 0.1785, 0.1689, 0.1977, 0.1764, 0.1882, 0.2066,
0.1976, 0.1820, 0.1959, 0.1923, 0.1890, 0.2039, 0.1843,
0.1626, 0.1936, 0.1497, 0.1687, 0.1552, 0.1930, 0.2002,
0.1945, 0.1496, 0.1902, 0.1932, 0.1964, 0.1960, 0.1302,
0.2041, 0.1938, 0.1172, 0.1964, 0.1644, 0.1392, 0.2242,
0.2161, 0.1988, 0.1831, 0.1781, 0.2014, 0.2313, 0.1659,
0.1729, 0.1504, 0.1237, 0.1461, 0.2231, 0.2200, 0.1365,
0.1267, 0.1903, 0.1304, 0.1314, 0.1488, 0.2102, 0.2209,
0.2202, 0.2071, 0.1849, 0.1693, 0.2050, 0.1968, 0.1972,
0.1638, 0.1653, 0.2203, 0.1555, 0.1691, 0.1945, 0.1849,
0.2019, 0.2369, 0.1930, 0.2014, 0.2247, 0.2038, 0.1653,
0.1384, 0.3183, 0.1976, 0.2010, 0.1879, 0.1703, 0.1880,
0.2127, 0.1852, 0.2193, 0.2023, 0.2166, 0.1547, 0.1343,
0.1245, 0.2063, 0.1767, 0.1484], device='cuda:0')), ('layer3.2.bn2.bias', tensor([ 0.1261, -0.0592, -0.1530, -0.0470, -0.1184, -0.1236, 0.1090,
-0.0293, -0.0089, -0.0829, -0.0502, -0.0755, -0.1208, -0.0099,
-0.1175, -0.0523, -0.1513, -0.1094, -0.1369, -0.0541, -0.0950,
-0.1112, 0.0632, 0.0442, -0.0875, -0.1306, -0.0721, -0.0688,
0.0597, -0.1114, -0.1094, -0.0986, -0.1523, -0.1476, -0.1077,
-0.1678, -0.0850, -0.0130, 0.0036, -0.1019, -0.0578, -0.1377,
-0.0667, -0.1334, -0.1261, -0.0940, -0.1323, -0.1400, -0.1320,
0.0218, -0.0023, -0.0984, -0.1066, -0.0318, -0.0892, -0.0049,
-0.1687, -0.0458, -0.0667, 0.0597, -0.1081, -0.0626, 0.0059,
0.0809, 0.1068, 0.0169, -0.1402, -0.0211, -0.0945, -0.0861,
-0.1509, -0.0913, -0.0681, -0.0479, -0.0934, -0.1006, -0.1083,
-0.1893, -0.0356, -0.1046, 0.0581, 0.0141, -0.1715, -0.0722,
-0.1160, 0.1053, -0.0325, -0.1663, -0.1044, -0.1551, -0.1544,
-0.2016, -0.1557, -0.0711, 0.0726, -0.0614, -0.1188, -0.0827,
-0.0001, -0.0473, -0.0773, -0.0592, -0.0785, -0.0534, -0.0847,
-0.1179, -0.1452, -0.1172, -0.1300, -0.1376, -0.1530, -0.0843,
-0.0920, -0.0855, 0.1771, 0.1697, -0.1238, -0.1649, 0.0508,
-0.0626, -0.1231, -0.1618, -0.0673, -0.1061, -0.1146, 0.0651,
-0.0486, 0.0356, -0.0439, -0.0775, -0.1106, -0.1380, -0.0860,
-0.0879, 0.0390, 0.0845, -0.0404, -0.0728, 0.0040, -0.1621,
-0.0895, -0.0345, -0.0582, 0.0985, -0.0677, -0.1493, -0.0908,
0.0902, -0.1058, -0.0888, -0.0291, -0.1475, -0.0818, -0.1098,
-0.0837, -0.1302, -0.1154, -0.0978, -0.0429, -0.1099, -0.1344,
-0.1137, -0.0440, -0.0506, -0.1452, -0.1070, -0.0797, -0.0914,
-0.0804, -0.0723, -0.1296, -0.0922, -0.0959, -0.1298, -0.0512,
-0.0195, -0.0682, 0.0165, -0.0518, -0.0142, -0.0713, -0.0857,
-0.1434, -0.0327, -0.0978, -0.0656, -0.1323, -0.0737, 0.1082,
-0.1328, -0.1087, 0.0707, -0.1221, -0.0268, -0.0098, -0.0983,
-0.1573, -0.1404, -0.0860, -0.0738, -0.0679, -0.1446, -0.0943,
-0.0618, 0.0091, 0.0540, 0.0458, -0.1048, -0.1252, 0.0623,
0.2188, -0.1045, 0.0056, 0.0210, 0.0219, -0.1138, -0.1230,
-0.1054, -0.0924, -0.1264, -0.0353, -0.1399, -0.0961, -0.0533,
-0.0229, -0.0244, -0.1230, 0.0004, -0.0940, -0.1104, -0.0959,
-0.1058, -0.0490, -0.1199, -0.1409, -0.1094, -0.1123, -0.0217,
0.0100, -0.2202, -0.1588, -0.1203, -0.1066, -0.0357, -0.0422,
-0.1826, -0.0714, -0.1066, -0.0994, -0.0816, -0.0014, 0.0262,
0.0881, -0.1068, -0.0539, -0.0139], device='cuda:0')), ('layer3.2.bn2.running_mean', tensor([ 0.0281, -0.1613, -0.0610, -0.0621, -0.1252, -0.0823, -0.1391,
-0.0538, -0.1233, -0.0373, 0.1260, -0.0536, -0.0825, -0.1068,
-0.0563, -0.0584, -0.0698, 0.1327, -0.0642, -0.0461, -0.0730,
0.0044, -0.0818, -0.0925, -0.0995, -0.1801, -0.1110, -0.1909,
-0.1355, -0.0825, -0.0454, -0.1027, -0.0908, -0.1873, -0.0708,
0.2133, -0.0576, -0.0513, -0.0333, -0.2353, -0.0464, -0.0067,
-0.0737, -0.1505, -0.1745, -0.0265, -0.1672, -0.0505, -0.1764,
-0.0182, -0.1305, 0.0624, -0.1142, -0.0643, -0.0850, -0.0719,
-0.0382, -0.0527, -0.0509, -0.0717, -0.0568, -0.0620, 0.0259,
-0.0363, -0.0319, -0.0321, -0.1295, -0.0037, -0.0703, -0.1159,
-0.0850, -0.0960, -0.0343, -0.0008, -0.0819, -0.0208, -0.0373,
-0.3319, 0.0509, 0.2654, 0.0294, -0.0157, -0.1210, -0.1059,
-0.0660, -0.1389, 0.0045, -0.0571, -0.0944, -0.0787, -0.1047,
-0.1056, -0.0218, -0.0531, 0.0907, -0.1328, -0.0473, -0.0472,
-0.0046, 0.0373, -0.0116, -0.0687, -0.0516, -0.0777, -0.0223,
0.0463, -0.0911, -0.0191, 0.0013, -0.0908, -0.0101, -0.0298,
-0.0837, 0.0016, -0.0881, 0.1037, -0.0523, -0.0895, -0.0383,
-0.1088, -0.1357, -0.0740, -0.0470, -0.0084, -0.0168, -0.1628,
-0.0882, 0.0422, -0.0487, -0.0985, -0.0568, -0.1579, -0.0961,
-0.0745, 0.0614, -0.1299, -0.0255, -0.1767, 0.0179, -0.0055,
-0.0886, -0.0164, -0.1016, 0.1208, -0.0671, -0.0321, -0.0153,
-0.0880, -0.1112, -0.0433, -0.0375, -0.0780, -0.0586, -0.0816,
-0.0794, -0.1193, -0.0692, -0.0512, 0.0419, -0.0196, -0.0799,
-0.0800, -0.0475, -0.0570, -0.0255, -0.0548, -0.0558, -0.1280,
-0.0477, -0.0036, -0.0568, -0.0592, -0.1163, -0.0966, -0.0546,
0.0091, -0.0530, 0.0341, -0.0012, -0.0515, -0.0352, -0.0268,
-0.1168, -0.0385, -0.0823, -0.0879, -0.0520, -0.0415, 0.0819,
-0.0813, 0.0468, 0.0506, -0.0947, -0.0529, -0.0622, -0.0673,
-0.0380, -0.0732, -0.1064, -0.0812, -0.0150, -0.0641, -0.0441,
-0.1041, -0.0789, -0.0642, 0.0225, -0.0401, -0.0859, -0.0432,
0.3423, -0.1050, -0.0162, -0.0874, -0.0722, -0.0261, -0.0397,
-0.2018, -0.0790, -0.0484, -0.0108, -0.0774, -0.1743, -0.0280,
-0.0447, -0.0971, -0.1503, -0.0964, 0.0022, 0.0125, -0.0499,
-0.0289, -0.1417, -0.0719, -0.0270, -0.1229, -0.1236, -0.0484,
-0.0703, -0.2353, -0.1251, -0.0673, -0.1042, 0.0420, -0.0565,
-0.0720, -0.0442, -0.0785, -0.1007, -0.0073, -0.0268, -0.1244,
-0.0798, -0.1133, 0.0155, -0.0555], device='cuda:0')), ('layer3.2.bn2.running_var', tensor(1.00000e-02 *
[ 0.9682, 1.1202, 1.4728, 0.6470, 2.0809, 1.6008, 1.9634,
0.9575, 0.9547, 1.7536, 1.3876, 1.1456, 1.7937, 0.8075,
1.3600, 1.7755, 0.9895, 1.3596, 1.9989, 0.9612, 1.9844,
2.1048, 1.2048, 1.5014, 1.3369, 2.5092, 1.0725, 1.2135,
1.5163, 2.0443, 1.0489, 3.7148, 1.6871, 3.1583, 1.5414,
1.3612, 2.0726, 0.9273, 1.3820, 3.2988, 1.8221, 1.2246,
0.8369, 1.1792, 0.9085, 1.7577, 2.7695, 1.1525, 1.6582,
1.0846, 0.9079, 0.7094, 0.9992, 1.0294, 1.1234, 3.1709,
1.2842, 1.2859, 1.5594, 1.0453, 2.6889, 0.9045, 1.2790,
1.5272, 1.7227, 0.8896, 1.5781, 1.0693, 1.0458, 1.2190,
0.9716, 0.9705, 0.9017, 1.1455, 1.8750, 1.9426, 1.4622,
1.1328, 1.9866, 1.6412, 1.6024, 0.8546, 1.7909, 0.9827,
1.1532, 2.1128, 1.3384, 0.8774, 1.2935, 1.2138, 1.4700,
1.0643, 2.0367, 1.7131, 2.6751, 0.9214, 0.9104, 1.7247,
1.7489, 1.3733, 1.7850, 1.5934, 0.9379, 1.6373, 1.8785,
1.3915, 1.6485, 1.6837, 0.6828, 1.3529, 2.0588, 1.1336,
1.2900, 0.7835, 1.2864, 1.8853, 2.1630, 1.6766, 1.1411,
1.2977, 2.3778, 1.1054, 0.9825, 0.8393, 0.9583, 1.1659,
1.7231, 1.2123, 1.2029, 1.2552, 1.6765, 2.4686, 0.8519,
1.2900, 1.5180, 1.2230, 1.5829, 1.7137, 1.5840, 1.7819,
1.3111, 1.4546, 1.9145, 1.0977, 0.8857, 1.4998, 1.1018,
1.7424, 1.4856, 1.0448, 2.8882, 1.4390, 2.3597, 1.1128,
1.1684, 0.9140, 2.4841, 1.1780, 2.1779, 1.8731, 2.7158,
1.3739, 1.4712, 1.0977, 0.8282, 0.8825, 1.3135, 3.0778,
1.8831, 1.2081, 1.2586, 2.0706, 1.0382, 1.1962, 1.5974,
1.6701, 1.8382, 1.2252, 1.2914, 1.6351, 1.4708, 1.7244,
1.1315, 0.8473, 1.3328, 0.9516, 0.8534, 1.1839, 2.3125,
1.4165, 0.9429, 1.2901, 1.1794, 0.8089, 0.8160, 3.2633,
1.5166, 1.1319, 0.8236, 1.4621, 1.8474, 2.0606, 1.2728,
1.0484, 1.3777, 1.2706, 1.5703, 2.5570, 2.2024, 1.2762,
1.6924, 1.1912, 1.4070, 1.2336, 1.2306, 1.3130, 1.7878,
2.0590, 1.7510, 1.0857, 1.3944, 1.7488, 1.1252, 2.0839,
1.5746, 1.0857, 1.6963, 0.8861, 0.8676, 1.9941, 1.2018,
2.2948, 2.9416, 1.7874, 1.2444, 1.8725, 2.0640, 1.0447,
1.2708, 3.0034, 1.8440, 0.9974, 1.4786, 1.0373, 1.7033,
1.4954, 1.7815, 2.1230, 1.6827, 1.6451, 1.4283, 1.0146,
1.3953, 1.3559, 1.9101, 1.6999], device='cuda:0')), ('layer3.2.conv3.weight', tensor([[[[-2.3780e-02]],
[[ 8.7015e-03]],
[[ 1.8315e-02]],
...,
[[-4.2013e-02]],
[[ 3.4015e-03]],
[[-9.9530e-03]]],
[[[-2.0450e-02]],
[[-1.2731e-02]],
[[-1.6573e-02]],
...,
[[ 1.6109e-02]],
[[-1.3517e-02]],
[[-1.6127e-02]]],
[[[ 1.0084e-02]],
[[ 2.4977e-03]],
[[-1.5499e-02]],
...,
[[-1.1960e-02]],
[[ 1.1546e-02]],
[[ 2.3375e-02]]],
...,
[[[-2.9875e-02]],
[[ 5.2212e-02]],
[[ 5.5405e-03]],
...,
[[-1.4726e-02]],
[[ 5.1513e-03]],
[[ 3.5456e-02]]],
[[[-1.1258e-02]],
[[-3.4418e-02]],
[[-1.6843e-03]],
...,
[[ 4.3885e-04]],
[[-1.5460e-02]],
[[ 1.8400e-02]]],
[[[ 2.4465e-02]],
[[ 1.4555e-05]],
[[-9.1697e-03]],
...,
[[ 1.8262e-02]],
[[-4.5732e-03]],
[[-1.8989e-02]]]], device='cuda:0')), ('layer3.2.bn3.weight', tensor([ 8.6847e-02, 8.2131e-02, 1.3081e-01, ..., 9.9649e-02,
5.0544e-02, 3.9904e-02], device='cuda:0')), ('layer3.2.bn3.bias', tensor([ 0.0686, -0.0592, -0.0134, ..., -0.0592, -0.0634, -0.0250], device='cuda:0')), ('layer3.2.bn3.running_mean', tensor([ 1.5163e-02, -1.6069e-02, -6.0626e-03, ..., 2.7971e-02,
-3.7536e-02, 4.2992e-03], device='cuda:0')), ('layer3.2.bn3.running_var', tensor([ 2.0756e-03, 4.6296e-04, 1.0436e-03, ..., 9.0061e-04,
5.6101e-04, 4.1729e-04], device='cuda:0')), ('layer3.3.conv1.weight', tensor([[[[-1.5120e-02]],
[[-9.6008e-03]],
[[-1.6491e-04]],
...,
[[-3.1888e-02]],
[[ 2.5355e-02]],
[[-2.2190e-02]]],
[[[ 6.7242e-03]],
[[ 1.3888e-03]],
[[ 3.9784e-03]],
...,
[[ 3.5906e-03]],
[[-2.0859e-02]],
[[ 7.3752e-03]]],
[[[ 7.9607e-03]],
[[ 2.0202e-03]],
[[ 1.1938e-02]],
...,
[[-1.4445e-02]],
[[-1.2397e-02]],
[[-8.0933e-03]]],
...,
[[[-1.4771e-02]],
[[-4.0543e-03]],
[[ 4.8153e-02]],
...,
[[ 3.5248e-02]],
[[ 4.6956e-03]],
[[ 2.8171e-02]]],
[[[-2.1148e-03]],
[[-7.5839e-03]],
[[ 2.1941e-02]],
...,
[[-7.1092e-03]],
[[-9.4924e-03]],
[[-1.0650e-02]]],
[[[ 9.0164e-03]],
[[ 1.3322e-02]],
[[-1.8748e-03]],
...,
[[ 1.1272e-02]],
[[-9.9430e-03]],
[[-6.4172e-04]]]], device='cuda:0')), ('layer3.3.bn1.weight', tensor([ 0.2229, 0.1737, 0.1807, 0.1743, 0.2112, 0.2004, 0.1866,
0.1774, 0.1899, 0.1719, 0.1934, 0.1197, 0.1730, 0.2153,
0.1575, 0.1363, 0.1404, 0.1833, 0.1611, 0.1692, 0.1554,
0.1915, 0.1046, 0.1651, 0.1605, 0.1914, 0.2084, 0.1711,
0.1402, 0.1871, 0.1471, 0.1801, 0.1525, 0.1472, 0.1256,
0.1489, 0.1293, 0.1581, 0.0954, 0.1441, 0.1945, 0.1885,
0.1622, 0.1384, 0.2003, 0.1829, 0.2135, 0.1420, 0.1324,
0.1230, 0.2077, 0.1245, 0.1603, 0.1811, 0.1482, 0.1650,
0.1502, 0.1305, 0.1799, 0.1917, 0.1686, 0.1494, 0.1098,
0.1833, 0.1073, 0.1492, 0.1856, 0.1065, 0.2008, 0.1309,
0.1700, 0.1822, 0.1689, 0.1534, 0.1665, 0.1794, 0.1505,
0.1328, 0.1693, 0.1629, 0.1650, 0.1527, 0.1627, 0.1938,
0.1708, 0.1397, 0.1746, 0.1271, 0.2141, 0.1936, 0.1817,
0.1516, 0.1131, 0.1471, 0.1978, 0.1723, 0.1694, 0.1051,
0.1875, 0.1729, 0.1972, 0.1727, 0.1708, 0.1723, 0.1200,
0.2013, 0.1295, 0.1470, 0.1613, 0.1778, 0.2323, 0.1622,
0.1757, 0.1291, 0.2197, 0.1441, 0.1044, 0.1629, 0.1796,
0.1346, 0.1685, 0.1939, 0.1750, 0.1967, 0.1755, 0.1417,
0.1717, 0.1709, 0.1698, 0.1879, 0.1836, 0.1870, 0.1532,
0.1133, 0.1326, 0.1823, 0.1561, 0.1784, 0.1598, 0.1860,
0.1816, 0.1350, 0.1126, 0.1335, 0.1228, 0.1490, 0.2010,
0.1749, 0.1740, 0.1583, 0.1481, 0.1624, 0.2132, 0.1403,
0.1671, 0.1582, 0.1300, 0.1463, 0.1615, 0.2289, 0.2171,
0.1641, 0.0988, 0.1869, 0.1842, 0.1168, 0.1693, 0.2242,
0.1005, 0.1915, 0.1709, 0.1660, 0.1837, 0.2081, 0.1536,
0.1580, 0.1237, 0.1597, 0.1689, 0.1735, 0.1842, 0.1718,
0.1431, 0.1978, 0.1284, 0.1829, 0.1815, 0.2245, 0.1361,
0.1437, 0.1653, 0.1587, 0.1626, 0.1660, 0.1746, 0.1734,
0.1873, 0.1886, 0.1831, 0.1484, 0.1438, 0.1444, 0.1328,
0.1518, 0.1265, 0.1693, 0.1622, 0.1627, 0.2238, 0.1581,
0.1725, 0.2309, 0.1473, 0.1756, 0.2244, 0.1528, 0.1398,
0.1826, 0.1346, 0.2191, 0.2090, 0.1422, 0.1609, 0.1321,
0.1493, 0.1340, 0.1712, 0.1931, 0.1718, 0.1942, 0.1771,
0.1857, 0.2283, 0.1482, 0.1387, 0.1013, 0.1606, 0.1327,
0.1716, 0.1277, 0.1832, 0.1371, 0.1285, 0.1129, 0.1627,
0.1624, 0.2447, 0.1932, 0.1969, 0.1170, 0.1720, 0.1099,
0.1518, 0.1755, 0.1738, 0.1695], device='cuda:0')), ('layer3.3.bn1.bias', tensor([-0.1493, -0.1098, -0.1351, -0.1251, -0.1526, -0.1701, -0.1150,
-0.1162, -0.0975, -0.0508, -0.1816, 0.0423, -0.0725, -0.1806,
-0.0772, -0.0545, -0.0323, -0.1313, -0.0705, -0.0686, -0.0902,
-0.1652, 0.0729, -0.1132, -0.1488, -0.1346, -0.1620, -0.1035,
-0.0289, -0.1819, -0.0022, -0.1459, -0.0465, -0.0622, -0.0050,
-0.0371, -0.0060, -0.0868, 0.0608, -0.0289, -0.1516, -0.1710,
-0.0588, -0.0257, -0.1779, -0.1216, -0.1475, -0.0150, 0.0263,
-0.0091, -0.0892, -0.0462, -0.0417, -0.1193, -0.0470, -0.0961,
-0.0475, 0.0049, -0.1421, -0.1069, -0.1048, -0.0306, 0.0331,
-0.1258, 0.0479, -0.0238, -0.1015, 0.0266, -0.1187, 0.0004,
-0.0962, -0.1173, -0.0677, -0.0691, -0.0827, -0.1355, -0.0343,
0.0071, -0.0839, -0.1333, -0.1169, -0.0489, -0.0440, -0.1690,
-0.0856, -0.0032, -0.0717, 0.0133, -0.0350, -0.1415, -0.1403,
0.0029, 0.0337, -0.0251, -0.1236, -0.1280, -0.0916, 0.0970,
-0.1121, -0.0810, -0.1883, -0.1045, -0.0731, -0.0749, 0.0001,
-0.1319, 0.0051, -0.0193, -0.1233, -0.0567, -0.1921, -0.0311,
-0.0915, 0.0667, -0.2421, -0.0052, 0.0807, -0.1208, -0.1202,
-0.0206, -0.1534, -0.1107, -0.1222, -0.1651, -0.1074, -0.0079,
-0.0767, -0.0923, -0.0478, -0.0925, -0.0668, -0.1157, -0.0595,
0.0106, 0.0069, -0.1168, -0.0846, -0.1390, -0.0946, -0.0974,
-0.0993, 0.0064, 0.0283, -0.0450, 0.0176, -0.0621, -0.1985,
-0.0639, -0.0938, -0.0822, -0.0751, -0.0804, -0.2168, -0.0401,
-0.0584, -0.0840, -0.0069, -0.0093, -0.1150, -0.2161, -0.2233,
-0.0365, 0.0641, -0.1025, -0.1237, 0.0387, -0.0812, -0.2594,
0.0589, -0.1672, -0.1029, -0.0724, -0.1084, -0.1673, -0.0826,
-0.0617, 0.0635, -0.0243, -0.1157, -0.1184, -0.1096, -0.1167,
-0.0362, -0.2068, 0.0518, -0.0868, -0.1299, -0.1225, -0.0028,
-0.0547, -0.1010, -0.0875, -0.0721, -0.0831, -0.1297, -0.1451,
-0.0961, -0.1315, -0.1254, -0.0628, -0.0598, -0.0460, -0.0526,
-0.0797, 0.0034, -0.1196, -0.0946, -0.0493, -0.1365, -0.0949,
-0.0641, -0.2518, -0.0419, -0.1462, -0.1635, -0.0351, -0.0299,
-0.1489, -0.0268, -0.1413, -0.2461, -0.0131, -0.0891, 0.0193,
-0.0712, 0.0130, -0.1153, -0.1656, -0.1328, -0.1235, -0.1661,
-0.1652, -0.1108, -0.0142, -0.0138, 0.0834, -0.1170, -0.0050,
-0.1065, 0.0409, -0.1423, 0.0520, -0.0144, 0.0022, -0.0853,
-0.0658, -0.2020, -0.0852, -0.1372, 0.1105, -0.1073, 0.0372,
-0.1313, -0.0850, -0.1663, -0.0795], device='cuda:0')), ('layer3.3.bn1.running_mean', tensor([-0.2620, -0.0437, -0.1546, -0.0487, -0.0700, -0.0164, -0.0925,
-0.0942, -0.0780, -0.1494, -0.1101, -0.1133, -0.0584, -0.0402,
-0.0753, -0.1066, 0.0108, -0.0866, -0.0399, -0.0243, -0.0040,
0.0214, -0.1489, -0.0932, -0.1256, -0.0453, -0.1288, -0.0556,
-0.0732, 0.0318, -0.1737, 0.0129, -0.0795, -0.0565, -0.1090,
-0.0735, -0.0287, -0.0456, -0.0137, -0.0744, -0.0285, 0.0717,
-0.0283, -0.1457, -0.0803, -0.0604, -0.1152, -0.0529, -0.1451,
-0.0548, -0.0356, -0.0611, -0.0840, -0.0753, -0.0642, -0.0559,
-0.0534, -0.1262, -0.0680, -0.1520, -0.0413, -0.0808, 0.0147,
-0.0356, -0.1137, -0.0464, -0.2355, -0.0299, -0.2247, -0.0549,
-0.0340, -0.0128, -0.0282, 0.0133, 0.1843, -0.0547, -0.0571,
-0.0489, -0.0655, -0.0396, 0.0141, 0.0048, 0.0127, -0.0728,
-0.0153, -0.1227, -0.1545, -0.1043, -0.2079, -0.1378, -0.1245,
-0.1507, -0.1649, -0.1129, -0.1633, -0.0370, -0.0007, -0.0898,
-0.1120, -0.1618, 0.0581, -0.0563, -0.0083, -0.0474, -0.0695,
-0.1902, 0.0342, -0.1289, -0.0665, -0.1100, 0.0858, -0.0241,
-0.1007, -0.0095, -0.1509, -0.1034, -0.0804, -0.0035, -0.0512,
-0.0500, -0.0310, -0.0773, -0.0646, -0.0250, -0.0004, -0.0265,
0.1267, -0.0037, -0.0250, -0.1475, -0.0934, -0.0755, -0.1045,
-0.1032, -0.1526, -0.0851, -0.0831, 0.0227, -0.1217, -0.1531,
-0.1233, 0.0676, 0.0037, -0.0668, -0.0511, -0.0228, -0.0812,
-0.0419, -0.0083, -0.0241, -0.0685, -0.0797, -0.0879, 0.0232,
-0.0219, -0.0141, -0.0560, 0.0809, -0.0067, -0.0230, 0.0232,
-0.0011, -0.1278, -0.0184, -0.0732, -0.2049, -0.0134, -0.0338,
-0.0295, -0.0514, -0.0846, 0.0108, -0.0788, -0.0826, -0.0727,
-0.0216, -0.0944, -0.1363, -0.0254, -0.1197, -0.0358, -0.0217,
-0.0908, -0.0410, -0.1140, -0.0948, -0.0562, -0.0755, -0.0898,
0.0441, -0.1051, -0.0657, -0.1155, -0.0213, 0.0060, 0.0713,
-0.1612, -0.0719, 0.0311, -0.0268, -0.0392, 0.0431, -0.0574,
-0.0292, -0.0422, -0.1526, -0.0154, 0.0079, -0.1812, -0.0248,
-0.1387, 0.0142, -0.0800, -0.0373, -0.0890, -0.0161, 0.0300,
-0.0539, -0.0833, -0.0850, -0.1196, -0.0443, 0.0176, -0.0661,
-0.1225, -0.1377, 0.0249, -0.0612, -0.0181, 0.0291, -0.0342,
-0.0823, -0.0925, -0.1396, -0.0467, 0.0080, 0.0057, -0.0021,
-0.0059, -0.1537, 0.0014, -0.1392, -0.1249, -0.0227, -0.0475,
-0.0341, -0.0861, -0.0775, -0.1406, 0.0035, -0.0430, -0.0115,
-0.0113, 0.0372, -0.0212, -0.0438], device='cuda:0')), ('layer3.3.bn1.running_var', tensor(1.00000e-02 *
[ 1.4987, 1.3058, 1.6935, 1.1079, 1.6192, 1.2820, 1.7705,
1.6238, 2.6320, 1.6659, 1.3875, 1.4222, 1.5619, 1.0570,
1.2235, 1.2096, 1.3292, 1.7671, 1.4103, 1.5599, 1.6728,
1.6438, 1.9448, 1.0741, 0.9984, 1.6916, 1.7431, 1.1703,
2.2265, 1.4716, 2.3364, 1.3681, 1.7376, 1.4853, 1.3290,
1.7963, 1.1776, 1.5914, 1.4997, 1.6314, 1.3149, 2.1451,
1.5325, 1.4216, 1.1936, 1.1708, 1.7126, 1.3570, 2.0419,
1.0052, 4.9452, 1.0889, 1.9821, 1.5279, 1.6928, 1.8354,
1.4448, 1.5237, 1.2222, 1.3433, 1.3548, 1.5179, 1.9706,
1.0440, 1.4857, 1.8362, 2.0476, 1.5072, 2.6264, 1.6510,
2.0049, 1.5129, 1.7975, 1.3750, 1.8014, 1.2835, 1.8330,
2.1310, 1.3954, 1.2603, 1.1042, 1.4635, 1.7869, 1.5066,
1.7657, 1.6979, 1.9170, 1.6598, 2.6273, 1.7277, 1.3309,
1.9093, 1.9564, 1.6634, 1.5593, 1.6484, 1.4803, 1.6621,
1.4057, 1.7502, 1.8685, 1.8490, 1.8193, 1.7410, 0.9858,
1.4931, 1.7224, 1.1224, 2.0402, 1.8693, 1.7717, 1.9130,
1.4500, 2.7726, 1.5300, 1.8681, 1.6488, 0.9141, 1.0431,
1.3718, 0.9391, 1.9945, 1.6484, 1.6191, 1.3293, 1.7752,
1.8148, 1.5530, 1.8184, 2.2269, 3.1586, 1.3245, 1.4744,
1.2895, 2.1011, 1.9377, 1.4060, 1.9808, 1.5824, 1.6825,
1.6891, 1.4819, 1.5551, 1.3854, 1.4389, 1.6187, 1.3600,
2.4801, 1.5957, 1.2611, 1.6586, 1.2681, 1.6291, 1.5730,
1.5639, 1.3534, 1.7347, 2.3241, 1.1707, 1.9866, 1.8825,
1.8510, 1.1264, 1.8031, 1.7362, 2.0827, 1.5058, 1.1273,
2.0178, 1.3443, 1.3440, 1.3920, 1.2193, 1.4185, 1.5829,
1.5732, 1.9791, 1.4126, 1.1594, 1.3698, 1.1694, 0.8912,
1.3780, 0.9441, 3.4400, 2.9302, 1.8124, 3.5386, 1.4016,
1.2015, 1.2018, 1.5529, 2.2860, 1.3889, 0.8743, 1.4684,
1.5709, 1.1813, 1.5727, 1.2279, 1.0655, 1.5240, 0.8446,
1.1616, 1.6219, 1.2845, 1.1852, 1.4820, 1.7629, 1.1707,
1.4748, 1.7535, 1.1631, 1.1094, 1.0800, 1.4316, 1.6248,
1.6351, 1.8502, 1.6687, 1.6346, 1.9717, 1.1091, 1.4157,
1.3577, 1.4796, 1.4372, 1.2912, 1.3492, 1.6553, 1.2177,
1.2373, 3.2880, 1.6697, 1.2226, 1.8329, 1.2218, 1.3635,
1.7246, 1.2315, 1.1028, 3.1395, 1.5512, 1.1666, 1.7189,
1.2218, 1.6817, 2.6300, 1.3158, 2.7987, 1.4087, 1.5562,
1.0822, 1.6459, 1.1145, 1.3967], device='cuda:0')), ('layer3.3.conv2.weight', tensor([[[[-2.0021e-02, -1.8551e-02, -1.7702e-02],
[ 3.1990e-03, 2.4608e-02, 8.0167e-03],
[ 6.3028e-04, 1.6046e-02, 1.4803e-02]],
[[-9.5785e-03, -2.7887e-02, -2.9613e-02],
[-3.1373e-02, -3.2521e-02, -3.0791e-02],
[-1.4816e-02, -2.8355e-02, -1.3990e-02]],
[[-1.3157e-02, -2.2484e-02, -1.0739e-02],
[-3.1500e-02, -2.3580e-02, 6.3404e-03],
[-2.5687e-02, -2.5521e-02, 4.3534e-03]],
...,
[[ 1.2378e-02, -1.2427e-04, 1.5254e-02],
[-6.0275e-04, -6.7721e-03, -2.6011e-03],
[-1.7559e-02, 1.8965e-02, -4.6312e-03]],
[[-3.6175e-03, -4.2797e-03, -1.7449e-03],
[-3.8641e-03, 7.3817e-04, -1.3511e-02],
[ 6.3750e-03, 1.6524e-02, 2.8192e-04]],
[[-3.1021e-03, -1.0760e-02, -1.0820e-02],
[ 7.6638e-03, 2.9162e-03, -5.7753e-03],
[ 1.1449e-02, 1.4833e-02, 9.0975e-03]]],
[[[ 1.5001e-02, 2.4205e-02, 1.2504e-02],
[ 3.4874e-03, -1.1496e-03, 3.4027e-03],
[ 2.7936e-05, 1.3407e-02, 1.1299e-02]],
[[-1.4379e-03, 7.5750e-03, 2.6118e-03],
[ 7.8706e-03, 1.8343e-02, 1.5806e-02],
[ 1.9598e-02, 1.7216e-02, 1.5445e-02]],
[[ 1.8630e-03, 1.0110e-02, 6.0079e-03],
[-1.5777e-03, 1.1053e-02, 1.2386e-02],
[ 3.2049e-04, 6.9257e-03, 1.6020e-02]],
...,
[[ 3.1639e-02, 5.4614e-02, 2.8195e-02],
[ 1.7428e-03, -5.5484e-03, 3.2060e-03],
[ 7.9264e-03, 1.1677e-02, -7.1334e-05]],
[[ 2.5534e-03, 2.4595e-02, 5.9373e-03],
[-1.7774e-02, -1.3784e-02, -2.2712e-02],
[-1.6822e-02, -2.6017e-02, -2.7600e-02]],
[[-1.0540e-02, 8.9218e-03, -1.3999e-02],
[-9.6549e-03, 2.3946e-02, -8.7016e-03],
[-7.9950e-03, 9.3906e-04, -9.1844e-03]]],
[[[ 1.0220e-02, 1.3448e-02, 3.1288e-03],
[ 1.5761e-02, 1.8226e-02, -6.3891e-03],
[-2.4425e-02, -1.9363e-02, -1.5872e-02]],
[[-3.9879e-03, 8.0336e-03, -2.0417e-02],
[ 8.3537e-04, 2.8709e-02, 6.7334e-03],
[ 7.4572e-03, 7.4434e-03, 1.9225e-03]],
[[ 4.3888e-03, -1.2994e-02, -2.3043e-02],
[ 2.7014e-02, 2.6314e-02, -2.5255e-02],
[-4.4141e-02, 2.8281e-02, 1.8219e-02]],
...,
[[ 2.1192e-02, 3.0951e-02, 1.8161e-02],
[ 2.0311e-02, -5.5866e-03, 8.4678e-03],
[ 1.8127e-02, 1.2721e-02, -1.8260e-02]],
[[ 5.3415e-04, 1.0669e-02, -1.4406e-03],
[-4.9864e-03, -3.8027e-03, -1.1200e-02],
[-6.7161e-03, -1.5710e-03, -5.0357e-03]],
[[ 5.3110e-03, -2.5468e-02, 4.4721e-03],
[ 1.7959e-02, 3.0402e-02, 2.5463e-02],
[ 5.8145e-03, 2.1388e-02, 6.7136e-03]]],
...,
[[[ 4.9446e-03, 1.7032e-02, -4.1625e-03],
[ 3.4180e-03, 1.2317e-02, -5.9132e-03],
[-8.0229e-03, 1.0697e-03, -4.7393e-03]],
[[-1.3029e-02, 4.4715e-03, 1.3844e-02],
[ 1.9024e-02, 6.9983e-03, -1.8130e-02],
[-8.5130e-03, -1.0460e-02, 5.6612e-03]],
[[-2.8004e-02, 1.6113e-03, -1.1320e-02],
[ 7.5326e-04, -1.1488e-03, 2.6678e-03],
[ 1.0549e-02, 1.7583e-02, 5.4505e-03]],
...,
[[ 3.1530e-03, -2.0338e-03, -6.5455e-03],
[ 1.5567e-03, 9.2643e-03, 1.6332e-03],
[ 1.0746e-03, 1.2560e-02, 3.5943e-02]],
[[-1.0374e-02, -7.9724e-03, 1.0137e-02],
[-1.2538e-03, -4.4716e-03, 1.2623e-02],
[ 7.9506e-03, -8.7204e-03, -4.3344e-03]],
[[ 9.0244e-03, -2.2564e-02, -1.3702e-02],
[-1.5199e-02, 2.4168e-03, -7.6975e-04],
[-1.2563e-03, -4.3446e-03, 1.3148e-02]]],
[[[-1.2275e-02, -1.2754e-02, -5.1172e-03],
[-1.1370e-03, -9.2372e-03, 8.6416e-04],
[ 4.9617e-03, -9.3581e-03, 3.2511e-03]],
[[-6.6198e-03, -6.9348e-03, -1.3156e-02],
[-5.2331e-03, -7.3249e-03, -1.1559e-02],
[-5.0624e-03, 9.6311e-03, -7.8505e-03]],
[[ 4.0814e-02, 7.9380e-03, -2.7049e-03],
[ 2.2294e-02, 9.4540e-03, -1.5283e-02],
[-6.1956e-03, -1.7091e-02, -3.0600e-02]],
...,
[[ 1.4390e-02, 1.5792e-03, 3.7335e-03],
[-1.1748e-02, -1.1045e-02, 1.5483e-03],
[ 1.5680e-02, 2.7082e-02, 1.0973e-02]],
[[-1.8610e-03, -1.1977e-02, 4.8226e-03],
[-1.4524e-03, -2.0644e-03, 3.1556e-03],
[-8.9026e-03, 2.0469e-03, -2.4178e-03]],
[[-1.4402e-02, -9.4209e-03, -1.2492e-02],
[-1.1095e-02, -9.3371e-03, -1.4696e-02],
[ 5.7585e-03, 1.5352e-02, -3.0816e-03]]],
[[[ 6.7708e-03, -2.6976e-02, 6.9671e-04],
[ 6.6295e-03, -2.3181e-02, -6.4667e-03],
[ 3.4421e-03, -1.4330e-02, 1.7412e-03]],
[[-2.3924e-02, -1.9910e-02, -1.2937e-02],
[-2.9762e-03, 6.6023e-03, -1.2093e-02],
[ 8.2873e-03, -1.0582e-02, 1.4183e-03]],
[[ 4.2226e-03, 1.5073e-03, 1.7971e-02],
[ 6.0556e-03, -5.8182e-03, 7.9520e-03],
[-3.6708e-03, -2.5598e-02, -1.9481e-02]],
...,
[[-5.5009e-03, -5.5731e-03, -4.2189e-03],
[ 3.2219e-04, -1.4452e-02, -1.0267e-02],
[-1.7291e-03, 6.5764e-04, 7.7291e-03]],
[[-3.4515e-03, -7.9305e-03, -7.1792e-03],
[-9.6562e-03, 7.9661e-03, 3.4565e-03],
[-1.2274e-03, 6.5865e-03, 1.6233e-03]],
[[ 8.3886e-03, 2.5827e-03, 1.8070e-04],
[-6.6978e-03, -6.0422e-03, 1.6910e-03],
[ 3.1690e-03, 6.6799e-03, 8.1717e-03]]]], device='cuda:0')), ('layer3.3.bn2.weight', tensor([ 0.1172, 0.1486, 0.1118, 0.1961, 0.1318, 0.1768, 0.2088,
0.1235, 0.1939, 0.1717, 0.1465, 0.1779, 0.1724, 0.1284,
0.2134, 0.1629, 0.1602, 0.1782, 0.1857, 0.1772, 0.2084,
0.1832, 0.1883, 0.2201, 0.1983, 0.1766, 0.1609, 0.1723,
0.1201, 0.1766, 0.1504, 0.1930, 0.1928, 0.1722, 0.1899,
0.1772, 0.1429, 0.1551, 0.1959, 0.1659, 0.1552, 0.1888,
0.1171, 0.1534, 0.1623, 0.1706, 0.1560, 0.1285, 0.1875,
0.1632, 0.2061, 0.1069, 0.1679, 0.1918, 0.1611, 0.1705,
0.1820, 0.1939, 0.1656, 0.2021, 0.2128, 0.2180, 0.2011,
0.1673, 0.1869, 0.1146, 0.1936, 0.1755, 0.1847, 0.1840,
0.1738, 0.1549, 0.1633, 0.2015, 0.1985, 0.1701, 0.1513,
0.1041, 0.2055, 0.1405, 0.1966, 0.1671, 0.1687, 0.1042,
0.1770, 0.1891, 0.2306, 0.1723, 0.2230, 0.1693, 0.1772,
0.1330, 0.1588, 0.1652, 0.1205, 0.1041, 0.0999, 0.1309,
0.1637, 0.1794, 0.1612, 0.1924, 0.1571, 0.1692, 0.1796,
0.2162, 0.1776, 0.2294, 0.1652, 0.1751, 0.1924, 0.1629,
0.1791, 0.2089, 0.1569, 0.2452, 0.1906, 0.1794, 0.1796,
0.1883, 0.1424, 0.1731, 0.1424, 0.1372, 0.1542, 0.1715,
0.1811, 0.1651, 0.1873, 0.1902, 0.1952, 0.2106, 0.1683,
0.1800, 0.1559, 0.1818, 0.1734, 0.1359, 0.1813, 0.1515,
0.1618, 0.1718, 0.1377, 0.1904, 0.1427, 0.1301, 0.1870,
0.1454, 0.2006, 0.1744, 0.1534, 0.1650, 0.1948, 0.1972,
0.1850, 0.1945, 0.1683, 0.2064, 0.1805, 0.2091, 0.1008,
0.1721, 0.1616, 0.1782, 0.1447, 0.1548, 0.1390, 0.1507,
0.1814, 0.1880, 0.1669, 0.1784, 0.2107, 0.1724, 0.1471,
0.1792, 0.1121, 0.1802, 0.1687, 0.1774, 0.1681, 0.1896,
0.1893, 0.1795, 0.1421, 0.2329, 0.1789, 0.1458, 0.1941,
0.1956, 0.2023, 0.1319, 0.1597, 0.2023, 0.1692, 0.1192,
0.1909, 0.1831, 0.2032, 0.1739, 0.1729, 0.2098, 0.1490,
0.1687, 0.1818, 0.2041, 0.1640, 0.1716, 0.1760, 0.2110,
0.1896, 0.1573, 0.2008, 0.1776, 0.1976, 0.1955, 0.1647,
0.1438, 0.1871, 0.1431, 0.1685, 0.1582, 0.1737, 0.2530,
0.1759, 0.1787, 0.1516, 0.1858, 0.2028, 0.1262, 0.1840,
0.1693, 0.1607, 0.1612, 0.1993, 0.0968, 0.1830, 0.1167,
0.1818, 0.1432, 0.1611, 0.1969, 0.1837, 0.1743, 0.1907,
0.1227, 0.1540, 0.2193, 0.1719, 0.1709, 0.1867, 0.1634,
0.2592, 0.1809, 0.1939, 0.1892], device='cuda:0')), ('layer3.3.bn2.bias', tensor([ 0.0455, 0.0114, 0.0543, -0.1035, -0.0039, -0.0757, -0.1422,
0.0248, -0.0766, -0.0710, -0.0464, -0.0480, -0.0542, 0.0290,
-0.1238, -0.0988, -0.0508, -0.0901, -0.0405, -0.0784, -0.1533,
-0.0721, -0.0873, -0.1499, -0.0973, -0.1233, -0.0495, -0.0790,
0.0652, -0.0384, -0.0393, -0.1157, -0.0972, -0.0794, -0.1821,
-0.0861, -0.0169, -0.0537, -0.1305, -0.0633, -0.0254, -0.0855,
0.0479, -0.0556, -0.0613, -0.0671, -0.0475, 0.0369, -0.1246,
-0.0789, -0.1112, 0.0760, -0.1156, -0.1098, -0.0638, -0.1069,
-0.0435, -0.1172, -0.0613, -0.1543, -0.1708, -0.1784, -0.0690,
-0.0399, -0.0932, 0.0419, -0.1154, -0.1171, -0.0923, -0.0639,
-0.1087, -0.1061, -0.0857, -0.0949, -0.1137, -0.0477, -0.0715,
0.0761, -0.1130, 0.0511, -0.0953, -0.0851, -0.0670, 0.1117,
-0.0071, -0.0763, -0.0939, -0.0684, -0.0762, -0.0408, -0.0951,
0.0644, -0.0695, -0.0269, 0.0201, 0.1716, 0.0915, 0.0008,
-0.0494, -0.0853, -0.0514, -0.1249, -0.0699, -0.0653, -0.0570,
-0.1052, -0.0793, -0.0728, -0.0819, -0.0286, -0.1101, -0.1126,
-0.0954, -0.1218, -0.0236, -0.1533, -0.1140, -0.0763, -0.0401,
-0.1015, -0.0260, -0.0764, -0.0094, 0.0063, -0.0123, -0.0924,
-0.0917, -0.0843, -0.1459, -0.1110, -0.1098, -0.1222, -0.0820,
-0.0868, -0.0236, -0.0188, -0.1086, -0.0003, -0.1303, -0.0408,
-0.0333, -0.0577, 0.0165, -0.1304, -0.0085, 0.0016, -0.0682,
0.0156, -0.1162, -0.0478, -0.0403, -0.0831, -0.1351, -0.1109,
-0.0897, -0.1121, -0.1022, -0.1640, -0.0827, -0.0951, 0.1404,
-0.0773, -0.0689, -0.1049, -0.0135, -0.0545, -0.0377, -0.0545,
-0.1116, -0.0349, -0.0096, -0.1127, -0.0501, -0.0897, -0.0169,
-0.1017, 0.1593, -0.0742, -0.1001, -0.0708, -0.0835, -0.1478,
-0.1208, -0.0753, 0.0439, -0.1303, -0.0392, -0.0390, -0.0879,
-0.1050, -0.1420, 0.0123, -0.0536, -0.1472, -0.1299, 0.0310,
-0.0642, -0.0952, -0.2083, -0.0673, -0.1081, -0.1043, -0.0352,
-0.0746, -0.1174, -0.0754, -0.1023, -0.0971, -0.0601, -0.1341,
-0.1156, -0.0317, -0.1022, -0.1406, -0.1641, -0.1076, -0.0631,
-0.0371, -0.1357, -0.0344, -0.0920, -0.0624, -0.0232, -0.1844,
-0.1186, -0.0680, -0.0322, -0.0491, -0.1180, 0.1064, -0.1022,
-0.1092, -0.0403, -0.0583, -0.0546, 0.1480, -0.1122, 0.0864,
-0.1407, 0.0219, -0.0551, -0.1320, -0.1212, -0.0852, -0.1310,
0.0837, 0.0098, -0.1358, -0.0812, -0.0854, -0.0734, -0.0591,
-0.0997, -0.0921, -0.1200, -0.1237], device='cuda:0')), ('layer3.3.bn2.running_mean', tensor([-0.0431, -0.0581, 0.0050, -0.1362, -0.1058, -0.0560, 0.0413,
0.0040, -0.1382, -0.0528, -0.0424, -0.0311, 0.0268, -0.0416,
-0.0407, -0.0495, -0.0723, -0.0220, -0.0096, -0.0436, -0.0889,
-0.1183, -0.0823, -0.1619, -0.1450, -0.0418, -0.0169, -0.1235,
-0.0552, -0.1035, -0.0900, -0.1083, -0.0247, 0.0206, 0.0105,
0.1650, 0.0313, 0.0306, -0.0606, -0.0509, -0.0912, -0.0853,
0.0814, 0.0039, -0.0347, -0.0008, -0.0837, -0.0001, -0.0340,
-0.0518, -0.1296, 0.0526, -0.0223, -0.0352, -0.0997, -0.0555,
-0.0768, -0.1072, -0.0611, -0.0808, -0.0751, -0.1046, -0.0739,
0.0303, -0.0654, 0.0257, -0.0343, -0.0986, -0.0772, -0.1252,
-0.0911, -0.1490, -0.0426, -0.1519, -0.1228, 0.0507, -0.0016,
-0.0236, -0.0576, -0.0362, -0.0800, -0.0795, 0.1248, 0.1732,
0.0379, -0.0645, -0.1627, -0.1218, -0.0917, -0.1208, -0.0071,
0.0276, 0.0300, -0.0184, -0.0724, 0.0173, -0.0843, -0.0978,
-0.0012, -0.0658, -0.0689, -0.0660, -0.0920, -0.1258, -0.0827,
-0.0550, -0.0668, -0.1048, -0.0669, -0.0153, -0.0016, -0.0834,
-0.0608, -0.0071, -0.1484, -0.0543, -0.1493, 0.0342, -0.0573,
0.0695, -0.0549, -0.1120, -0.0349, -0.1081, 0.0137, -0.0424,
-0.0857, 0.3789, -0.0199, -0.0409, -0.1016, -0.0266, -0.0448,
-0.0769, -0.1480, 0.0109, -0.0236, -0.0049, -0.0564, -0.0203,
-0.0742, -0.0124, -0.0948, 0.0131, 0.0563, -0.0918, 0.0171,
-0.1306, -0.0509, -0.0978, -0.0191, -0.1235, -0.0668, -0.0447,
-0.0809, -0.1201, -0.0365, -0.0978, -0.1199, -0.0733, 0.0476,
0.0507, -0.0704, -0.0565, -0.0497, -0.0292, -0.0697, -0.0696,
-0.0318, 0.0585, -0.1082, -0.0527, -0.0704, -0.0867, -0.1431,
-0.0817, -0.0711, -0.0264, -0.0011, -0.0821, -0.0234, -0.1078,
-0.0772, -0.1349, -0.1194, -0.1055, -0.1126, -0.0404, -0.0450,
-0.0369, 0.1679, -0.0673, -0.0868, -0.1119, -0.0587, -0.0075,
0.0128, -0.0449, -0.0892, -0.0801, -0.0507, -0.0871, 0.0148,
-0.0485, -0.1587, -0.0729, -0.0654, -0.0539, -0.1019, 0.0659,
-0.0173, -0.0131, -0.1204, -0.0178, -0.0896, -0.1189, -0.0916,
-0.0291, -0.0924, -0.0700, -0.0972, 0.0127, -0.0946, -0.0380,
-0.0928, -0.0436, -0.0578, 0.0058, -0.1299, 0.1045, -0.0413,
-0.0402, -0.0925, -0.1468, -0.0698, 0.0486, -0.1251, -0.0671,
-0.0702, -0.0746, -0.0637, -0.0223, -0.0893, -0.0056, -0.1242,
0.0417, -0.0586, -0.0222, -0.0587, 0.0083, -0.0185, -0.1425,
-0.1028, -0.0813, -0.0479, -0.0769], device='cuda:0')), ('layer3.3.bn2.running_var', tensor(1.00000e-02 *
[ 0.6158, 1.2463, 0.8077, 1.1543, 0.6859, 0.5870, 1.5879,
1.0907, 1.5306, 0.7470, 0.7210, 0.7859, 0.8278, 1.0818,
0.9813, 0.8284, 0.6440, 0.9558, 1.9072, 1.0265, 0.8654,
1.2493, 1.4135, 1.0231, 0.9605, 0.5405, 0.7819, 1.4482,
0.7991, 1.1677, 1.0343, 1.0349, 0.8740, 0.6556, 1.1777,
0.9516, 1.4753, 0.8919, 0.8388, 0.7423, 0.9494, 1.1932,
1.6877, 0.7341, 0.8169, 0.7491, 0.6635, 1.2184, 0.8436,
0.5852, 0.8603, 1.0916, 0.7326, 0.9760, 0.6723, 0.6089,
1.1328, 0.6970, 0.8924, 0.7228, 0.7947, 0.8838, 1.0929,
0.7477, 1.0376, 1.5907, 0.9872, 0.9965, 0.7908, 0.9951,
0.6849, 0.6435, 0.6009, 1.2493, 0.9506, 0.9333, 0.6158,
0.7435, 0.8575, 1.3657, 1.3712, 0.7661, 1.2016, 0.9434,
1.2075, 0.7283, 1.1743, 0.9523, 2.9163, 0.9026, 0.8328,
1.1630, 0.8669, 0.8695, 1.0928, 1.2045, 0.7740, 0.8679,
1.1048, 1.1879, 0.7073, 0.8264, 0.7451, 0.5786, 0.8792,
2.4741, 1.0227, 4.4171, 0.7306, 1.6977, 0.9927, 0.7005,
1.0412, 1.5072, 0.7969, 1.4505, 0.9054, 0.8463, 0.9219,
1.1803, 0.7852, 0.9555, 0.8418, 0.7700, 0.6665, 0.9236,
0.8284, 1.7347, 0.6409, 0.9956, 0.5844, 0.8989, 0.9372,
1.0725, 0.8176, 1.9325, 0.8049, 0.8218, 0.8348, 1.0255,
0.7364, 0.7043, 0.8163, 0.7580, 1.5320, 0.8463, 0.8252,
1.0638, 0.9256, 0.8980, 1.2886, 0.6869, 0.7115, 0.7507,
1.4048, 1.0351, 0.7539, 1.0321, 0.8670, 2.6504, 1.9455,
1.2763, 0.4835, 0.9011, 0.7371, 1.0810, 1.0464, 0.6483,
0.7992, 1.0356, 1.2336, 0.7712, 2.8628, 0.6851, 0.8951,
0.7821, 0.9505, 1.2188, 1.2774, 0.7356, 0.8401, 0.8026,
0.9773, 1.1546, 1.1627, 0.8805, 1.1529, 0.6667, 0.9718,
0.5943, 1.7344, 0.9150, 1.2689, 1.1916, 0.7361, 1.4454,
1.3005, 0.7778, 0.7932, 0.9110, 0.9887, 1.0503, 0.7176,
0.6621, 1.0051, 1.5242, 0.5835, 0.7467, 0.9755, 1.7089,
0.9522, 1.1004, 1.2726, 0.5807, 0.7997, 0.6799, 0.8894,
0.7977, 0.7214, 0.7088, 0.5735, 0.7778, 0.7736, 1.4516,
0.6981, 0.9676, 0.7986, 1.6922, 1.0735, 1.0632, 0.8969,
0.5714, 1.3183, 0.8113, 1.1412, 0.8028, 1.1048, 0.9895,
0.6780, 0.9466, 0.7854, 1.0222, 0.9075, 0.7828, 0.9006,
1.7309, 0.9630, 1.3031, 1.1606, 0.7599, 1.0640, 0.8162,
2.8563, 0.9494, 0.9030, 0.6846], device='cuda:0')), ('layer3.3.conv3.weight', tensor([[[[ 3.1927e-03]],
[[ 1.4805e-02]],
[[-3.9558e-02]],
...,
[[-1.3219e-03]],
[[ 1.5086e-02]],
[[-1.1912e-02]]],
[[[-5.6582e-03]],
[[ 2.2918e-02]],
[[-2.6290e-02]],
...,
[[ 3.1419e-02]],
[[-2.5987e-02]],
[[-1.2404e-03]]],
[[[-1.3780e-03]],
[[-3.3550e-03]],
[[ 1.2346e-02]],
...,
[[ 2.9028e-02]],
[[ 8.1570e-03]],
[[ 9.2157e-04]]],
...,
[[[ 9.3495e-03]],
[[-3.9317e-02]],
[[-1.5141e-03]],
...,
[[ 9.5364e-03]],
[[-2.4709e-02]],
[[ 3.2996e-03]]],
[[[-5.9284e-03]],
[[ 1.8051e-02]],
[[-4.4164e-02]],
...,
[[-7.3454e-03]],
[[ 7.7065e-03]],
[[-7.1259e-03]]],
[[[-7.3073e-03]],
[[-1.1021e-02]],
[[-1.2927e-02]],
...,
[[ 1.1101e-03]],
[[-6.2908e-03]],
[[-5.6577e-03]]]], device='cuda:0')), ('layer3.3.bn3.weight', tensor([ 3.6002e-02, 1.3361e-01, 1.0711e-01, ..., 1.1089e-01,
6.0103e-02, 1.6406e-02], device='cuda:0')), ('layer3.3.bn3.bias', tensor([ 0.0043, -0.0521, -0.0891, ..., -0.0850, -0.0222, -0.0078], device='cuda:0')), ('layer3.3.bn3.running_mean', tensor(1.00000e-02 *
[-2.1853, -3.0616, -3.7836, ..., -1.8853, -0.2141, 0.7410], device='cuda:0')), ('layer3.3.bn3.running_var', tensor([ 6.0664e-04, 1.1630e-03, 8.5144e-04, ..., 8.5821e-04,
6.6438e-04, 1.7357e-04], device='cuda:0')), ('layer3.4.conv1.weight', tensor([[[[-3.4127e-02]],
[[ 3.2188e-02]],
[[-2.0301e-02]],
...,
[[ 4.1546e-03]],
[[ 1.2702e-02]],
[[-5.6641e-03]]],
[[[ 3.2940e-03]],
[[-9.8673e-03]],
[[-8.8082e-03]],
...,
[[-1.5857e-03]],
[[ 7.6469e-03]],
[[-5.9873e-03]]],
[[[-2.4149e-02]],
[[ 1.2970e-02]],
[[-6.4782e-03]],
...,
[[ 1.3755e-02]],
[[ 1.3282e-02]],
[[ 1.6999e-04]]],
...,
[[[-1.1186e-02]],
[[-4.2158e-03]],
[[ 2.9875e-02]],
...,
[[ 1.6177e-02]],
[[ 7.1540e-03]],
[[ 5.6421e-03]]],
[[[ 4.7531e-03]],
[[-2.9031e-03]],
[[-6.5355e-03]],
...,
[[ 2.8871e-02]],
[[-1.5105e-02]],
[[-1.3327e-03]]],
[[[-1.2044e-02]],
[[-1.7834e-03]],
[[-1.6628e-02]],
...,
[[-1.4216e-02]],
[[-1.5398e-02]],
[[ 1.6957e-03]]]], device='cuda:0')), ('layer3.4.bn1.weight', tensor([ 0.1289, 0.1811, 0.1434, 0.1497, 0.1642, 0.2092, 0.1794,
0.1881, 0.1535, 0.1468, 0.1609, 0.1833, 0.1200, 0.1987,
0.1594, 0.1837, 0.2195, 0.1638, 0.1447, 0.2000, 0.1600,
0.1625, 0.1672, 0.1554, 0.1694, 0.1334, 0.1867, 0.1898,
0.1418, 0.1985, 0.2040, 0.1359, 0.2024, 0.1679, 0.1409,
0.1710, 0.1171, 0.1267, 0.1705, 0.1479, 0.1864, 0.1502,
0.2114, 0.2251, 0.1741, 0.1688, 0.1057, 0.1641, 0.1542,
0.1765, 0.1358, 0.1193, 0.1366, 0.1934, 0.1655, 0.1889,
0.1728, 0.1615, 0.1355, 0.1368, 0.1782, 0.1291, 0.1392,
0.1907, 0.1541, 0.1765, 0.1072, 0.1467, 0.1842, 0.1374,
0.1605, 0.1637, 0.1314, 0.1463, 0.2017, 0.1425, 0.1926,
0.1525, 0.1305, 0.1450, 0.1296, 0.1885, 0.1971, 0.1925,
0.1545, 0.1706, 0.2379, 0.2122, 0.1125, 0.1800, 0.1602,
0.2117, 0.1842, 0.1107, 0.1372, 0.2442, 0.1042, 0.1182,
0.0949, 0.1530, 0.2418, 0.1407, 0.1725, 0.1684, 0.1633,
0.1663, 0.1553, 0.1665, 0.1538, 0.1521, 0.1838, 0.1724,
0.1525, 0.1191, 0.2689, 0.2067, 0.1609, 0.2066, 0.1505,
0.1249, 0.1743, 0.1818, 0.1720, 0.1662, 0.1839, 0.1219,
0.1762, 0.1704, 0.2420, 0.1757, 0.1415, 0.1631, 0.1480,
0.1309, 0.1715, 0.2615, 0.2070, 0.1373, 0.1571, 0.1903,
0.2086, 0.1327, 0.1527, 0.1692, 0.1480, 0.1080, 0.2477,
0.1751, 0.1824, 0.1472, 0.1249, 0.1503, 0.1501, 0.1819,
0.1820, 0.1727, 0.1731, 0.2000, 0.1898, 0.1394, 0.1588,
0.2024, 0.1223, 0.1661, 0.1468, 0.1820, 0.1327, 0.1352,
0.1986, 0.1774, 0.1645, 0.1668, 0.1750, 0.1990, 0.1404,
0.1675, 0.1671, 0.1795, 0.2062, 0.2018, 0.1402, 0.1090,
0.2003, 0.1751, 0.2018, 0.1448, 0.2129, 0.1697, 0.1422,
0.1965, 0.2124, 0.1867, 0.1635, 0.1548, 0.1512, 0.1250,
0.1871, 0.1386, 0.1309, 0.1906, 0.1693, 0.2194, 0.1601,
0.1940, 0.1565, 0.1534, 0.1686, 0.1828, 0.1698, 0.1652,
0.1527, 0.1592, 0.2534, 0.1596, 0.1241, 0.1393, 0.1909,
0.1468, 0.1454, 0.1538, 0.1681, 0.2015, 0.1332, 0.1100,
0.1251, 0.1419, 0.1402, 0.1850, 0.1382, 0.1208, 0.1329,
0.1655, 0.2290, 0.2108, 0.1896, 0.2278, 0.1625, 0.1533,
0.1672, 0.1810, 0.1588, 0.1610, 0.1820, 0.1441, 0.1734,
0.1845, 0.1972, 0.1542, 0.1601, 0.1929, 0.1244, 0.1860,
0.1983, 0.1758, 0.1489, 0.2357], device='cuda:0')), ('layer3.4.bn1.bias', tensor([ 0.0006, -0.0725, -0.0153, -0.0658, -0.0610, -0.2592, -0.1810,
-0.1490, -0.0697, -0.0946, -0.1224, -0.1609, 0.0562, -0.0820,
-0.1319, -0.1549, -0.1721, -0.0335, -0.0094, -0.1652, -0.1359,
-0.1245, -0.0630, -0.0685, -0.1001, -0.0173, -0.1559, -0.1228,
-0.0454, -0.1486, -0.1025, -0.0473, -0.1797, -0.1243, -0.0262,
-0.0965, -0.0025, 0.0196, -0.0636, -0.0974, -0.2064, -0.0293,
-0.1324, -0.1492, -0.1018, -0.1239, 0.0444, -0.0776, -0.1004,
-0.1010, -0.0011, -0.0309, -0.0339, -0.1574, -0.0913, -0.1674,
-0.0259, -0.0579, -0.0208, -0.0215, -0.1362, 0.0058, -0.0442,
-0.1441, -0.0608, -0.0837, 0.0099, 0.0036, -0.1816, -0.0252,
-0.1104, -0.0940, -0.0362, -0.0573, -0.1263, -0.0292, -0.1641,
-0.0536, 0.0529, -0.0321, -0.0257, -0.0947, -0.1793, -0.1640,
-0.0930, -0.1015, -0.1547, -0.1717, 0.0330, -0.1644, -0.0914,
-0.1588, -0.1597, 0.0845, -0.0404, -0.2776, 0.0810, -0.0329,
0.1247, -0.1345, -0.3245, -0.0499, -0.1568, -0.1060, -0.0913,
-0.0739, -0.0661, -0.1299, -0.0879, -0.0423, -0.1727, -0.1591,
-0.0041, 0.0004, -0.3203, -0.0862, -0.0757, -0.1353, -0.0466,
0.0621, -0.1229, -0.1409, -0.1227, -0.1133, -0.0855, -0.0182,
-0.1056, -0.0657, -0.1713, -0.1127, -0.0370, -0.0839, -0.0119,
0.0098, -0.1237, -0.2439, -0.0838, -0.0089, -0.0526, -0.0859,
-0.1498, -0.0226, -0.0194, -0.1387, -0.0418, 0.0877, -0.3100,
-0.1448, -0.1051, -0.0301, -0.0122, -0.0606, -0.0785, -0.1823,
-0.0825, -0.1156, -0.1209, -0.2377, -0.1006, 0.0264, -0.0861,
-0.1095, 0.0006, -0.0606, -0.0640, -0.1517, 0.0061, -0.0355,
-0.1887, -0.0947, -0.1044, -0.1002, -0.0760, -0.1056, 0.0033,
-0.0441, -0.0809, -0.0876, -0.1879, -0.1663, -0.0124, 0.0668,
-0.1649, -0.1384, -0.2013, -0.0407, -0.1811, -0.0985, -0.0375,
-0.0793, -0.1937, -0.1418, -0.1152, -0.0526, -0.0784, -0.0468,
-0.1456, -0.0240, -0.0382, -0.1198, -0.1637, -0.2104, -0.1062,
-0.2088, -0.1195, -0.0362, -0.1108, -0.1116, -0.1301, -0.0474,
-0.0746, -0.0762, -0.2479, -0.0820, 0.0135, -0.0572, -0.1498,
-0.0102, -0.0639, -0.0599, -0.1032, -0.1265, -0.0680, 0.0054,
-0.0025, 0.0403, -0.0209, -0.1277, -0.0383, 0.0797, 0.0019,
-0.1176, -0.1810, -0.2023, -0.1025, -0.1213, -0.0214, -0.0593,
-0.1143, -0.1140, -0.1222, -0.0670, -0.1253, -0.0126, -0.0788,
-0.1192, -0.1596, -0.0995, -0.0627, -0.0890, 0.0575, -0.1436,
-0.1016, -0.1194, -0.0882, -0.1911], device='cuda:0')), ('layer3.4.bn1.running_mean', tensor([-0.0586, -0.0572, -0.1097, -0.0355, -0.1139, -0.0223, -0.1201,
-0.1030, 0.0096, -0.0588, 0.0454, -0.0309, -0.1623, -0.1432,
-0.0830, -0.0441, -0.0977, -0.1405, -0.0447, -0.1310, -0.0746,
-0.0121, -0.0889, -0.1681, 0.0488, -0.0240, -0.0223, -0.0208,
-0.0323, -0.0896, -0.1790, -0.0625, -0.0794, -0.0113, -0.0458,
-0.0054, -0.0790, -0.0430, -0.0235, 0.0445, -0.0568, -0.0832,
-0.1725, -0.1317, -0.0395, 0.1170, -0.0156, -0.0934, -0.1276,
-0.1655, -0.0211, -0.0062, -0.0839, -0.0580, -0.0756, -0.0564,
-0.0677, -0.0244, -0.0172, -0.0460, 0.0346, -0.1189, -0.1802,
-0.0680, -0.1142, -0.0978, -0.1274, -0.0406, -0.0713, -0.0357,
-0.0878, -0.0801, -0.0365, 0.0242, -0.0092, -0.0230, -0.0890,
-0.0544, -0.1220, -0.0390, -0.0563, -0.0579, -0.0886, -0.0133,
-0.0076, -0.0851, -0.0714, -0.1200, -0.0703, -0.0215, -0.0006,
-0.0609, 0.0117, 0.0419, -0.0873, -0.1451, -0.2438, -0.0571,
-0.2300, -0.0790, -0.1471, -0.0924, -0.0339, -0.0637, -0.0169,
-0.0634, -0.0907, -0.0378, 0.0454, -0.1111, -0.0274, -0.0703,
0.0213, -0.1324, -0.1548, -0.1171, -0.1212, -0.0172, 0.0609,
-0.0668, -0.0531, -0.1235, -0.0434, -0.0778, -0.0183, -0.1124,
-0.0152, -0.0100, 0.0702, 0.0315, -0.0861, -0.0230, -0.0971,
-0.1756, 0.0070, -0.1204, -0.1586, -0.0482, -0.0869, -0.1411,
-0.0673, -0.0299, -0.1318, -0.0547, -0.0812, -0.0894, -0.2307,
-0.0853, -0.0983, -0.0566, -0.0213, -0.0999, -0.0382, -0.0396,
-0.0682, 0.0033, -0.0735, -0.1005, -0.0891, -0.0129, 0.0095,
-0.0670, -0.0144, -0.1125, -0.0694, -0.1345, -0.1103, -0.1089,
-0.0080, -0.1107, -0.0117, -0.0697, -0.0574, -0.0847, -0.0416,
-0.1142, -0.1093, -0.0063, -0.1060, -0.1655, -0.1148, 0.0536,
-0.0696, 0.0402, -0.0001, -0.0422, 0.0832, 0.0125, -0.1150,
-0.0129, -0.1257, -0.0173, -0.0328, -0.0673, -0.0544, -0.0574,
-0.0465, -0.1242, -0.0245, -0.1465, -0.0757, -0.0748, -0.0175,
-0.0762, -0.0183, -0.0799, -0.1015, -0.1282, 0.0111, -0.0981,
-0.0203, -0.0450, -0.2117, -0.0139, -0.1421, 0.0363, -0.1493,
-0.1129, 0.0167, -0.0782, -0.0143, -0.0675, -0.0706, -0.0303,
-0.1068, -0.0797, -0.1899, 0.0009, -0.0664, -0.0481, -0.1535,
-0.0714, -0.0987, -0.1312, -0.1425, -0.1406, -0.1240, -0.0126,
-0.1287, -0.0448, -0.0465, -0.0289, -0.0572, -0.0356, -0.1932,
-0.1118, -0.1289, -0.0036, -0.0354, -0.1478, -0.1280, -0.0326,
-0.0819, -0.0285, 0.0210, -0.1554], device='cuda:0')), ('layer3.4.bn1.running_var', tensor(1.00000e-02 *
[ 1.4481, 1.0382, 1.6103, 2.3134, 1.4353, 0.9337, 1.3544,
1.6947, 1.2163, 1.7228, 1.2086, 1.2998, 1.6005, 2.2463,
1.0837, 1.1569, 1.6445, 1.8233, 1.4466, 1.1070, 1.2094,
0.9119, 2.1202, 1.2258, 1.3245, 1.8623, 1.0126, 1.2441,
1.0569, 1.5204, 2.4192, 1.6252, 1.1474, 1.4586, 1.4546,
1.2872, 1.2580, 1.0400, 1.4762, 1.1909, 1.0854, 1.6022,
2.1772, 1.7608, 1.2575, 1.1094, 1.1643, 1.2116, 1.0059,
1.1393, 1.3221, 1.4511, 1.2854, 1.0535, 1.1962, 1.2866,
2.6445, 2.0210, 1.4924, 1.2199, 0.9713, 1.7599, 1.5476,
1.8745, 1.7330, 1.4228, 1.6453, 2.0633, 1.0101, 1.4947,
1.2312, 1.8664, 1.1147, 0.9600, 1.8148, 1.5964, 1.1404,
1.0961, 2.8729, 1.3519, 1.0761, 1.1855, 1.1441, 1.3703,
0.8395, 1.1924, 1.4266, 1.7796, 1.4225, 1.1123, 1.2049,
1.4236, 0.9784, 1.6494, 1.7231, 1.6017, 2.1082, 1.0882,
1.5689, 1.0255, 1.1974, 1.7452, 1.2370, 1.3918, 0.9922,
1.0676, 1.4236, 1.1465, 1.2950, 1.0809, 1.1731, 1.4349,
2.7570, 1.5580, 1.7009, 1.4718, 1.2030, 1.4013, 1.3634,
2.6460, 2.2682, 1.3891, 0.9731, 1.1641, 1.7519, 1.1430,
2.1940, 1.3898, 3.1634, 1.3955, 1.2814, 1.3195, 1.2168,
1.6355, 1.2792, 1.5910, 2.1561, 2.2098, 2.0383, 2.0587,
1.5167, 1.0021, 2.2399, 1.1605, 1.4148, 1.9495, 1.5588,
1.8175, 1.4585, 1.5353, 1.8178, 1.2544, 1.1453, 0.9076,
1.4697, 1.2908, 1.3458, 0.9190, 1.4490, 2.2375, 1.2043,
1.6302, 1.1826, 1.2447, 1.2584, 1.6592, 1.8198, 1.2789,
0.9336, 1.6805, 1.5848, 1.0510, 1.5223, 2.0929, 1.7784,
1.8375, 1.3951, 1.7018, 1.0454, 1.3254, 1.9863, 1.6230,
1.0976, 1.5341, 1.3515, 1.0591, 1.9400, 1.5467, 1.6195,
1.9529, 2.0170, 0.9281, 1.0571, 1.1397, 1.2153, 1.6218,
1.6227, 1.1093, 1.1582, 1.4611, 0.9490, 1.2872, 1.3674,
0.9510, 0.9833, 1.6236, 2.4352, 1.3737, 1.2872, 1.8647,
1.2516, 1.4217, 1.3307, 1.0688, 1.7869, 1.2014, 1.6510,
1.7883, 1.0307, 1.6949, 1.1544, 1.5755, 1.0341, 1.4147,
2.4627, 2.4364, 1.5626, 1.8539, 1.3044, 2.2880, 1.7732,
0.8852, 1.5397, 1.3467, 1.7945, 1.9392, 1.6320, 1.4408,
1.1687, 1.1511, 1.5431, 1.1387, 1.1415, 1.1492, 2.2703,
1.1243, 1.5238, 1.3241, 2.1953, 2.2106, 2.6231, 1.3083,
1.6619, 1.0624, 0.8317, 2.6737], device='cuda:0')), ('layer3.4.conv2.weight', tensor([[[[-8.8975e-03, 8.2681e-03, 9.5468e-03],
[-2.9289e-02, -8.9024e-03, -5.8257e-03],
[-1.3357e-02, -2.4827e-02, -2.0100e-02]],
[[-2.0383e-05, -2.2157e-04, 2.0560e-03],
[-8.4056e-03, -2.0170e-02, -9.2057e-03],
[ 6.3248e-03, 2.9406e-03, 1.7933e-03]],
[[-2.3525e-02, -1.0879e-02, 1.3070e-02],
[-5.3119e-04, -2.0994e-02, -8.2173e-03],
[ 1.7326e-03, -1.0037e-02, -1.5591e-03]],
...,
[[ 3.6863e-03, -5.2400e-03, -4.0052e-03],
[-5.1566e-03, 1.3139e-02, 1.6186e-02],
[-1.3664e-02, 6.3976e-03, 1.2014e-04]],
[[ 3.8689e-03, -2.4797e-02, -4.8346e-03],
[-1.1029e-03, -7.9848e-03, -1.4444e-02],
[ 4.1494e-03, 1.7180e-02, 9.5370e-03]],
[[-1.2932e-04, -4.8040e-03, -5.6616e-03],
[-1.0475e-02, -9.0452e-03, -1.4027e-02],
[-4.2955e-03, -5.5446e-03, -5.4631e-03]]],
[[[-2.7454e-02, -2.0153e-02, -1.8434e-02],
[ 7.4258e-03, 2.4594e-02, 1.1151e-02],
[-6.7909e-03, 7.0744e-03, 2.1112e-02]],
[[-4.6547e-03, 8.3668e-03, -1.7892e-03],
[ 6.8007e-03, 1.5259e-02, 9.2129e-03],
[-9.3272e-03, 2.7739e-02, 2.1659e-02]],
[[ 9.6779e-03, 1.1780e-02, 3.2909e-02],
[-1.2889e-02, -5.5071e-03, 7.3504e-03],
[ 8.8885e-03, -1.7894e-02, -2.8907e-02]],
...,
[[ 1.2088e-02, 2.2231e-02, 3.1916e-02],
[-4.9173e-03, 7.6595e-03, -5.7592e-04],
[-7.4764e-03, -6.3305e-03, 8.6860e-03]],
[[ 1.1143e-02, 2.3442e-02, 2.0811e-02],
[-4.2744e-03, -8.3543e-03, -9.7011e-03],
[-1.4505e-02, -1.7824e-02, -2.2806e-02]],
[[-1.3286e-02, -2.2064e-02, -1.9249e-02],
[-3.6862e-03, -1.6956e-02, 1.0500e-03],
[-7.9056e-03, -1.2488e-02, -3.9062e-03]]],
[[[ 1.5650e-02, 5.5990e-03, 5.8535e-03],
[ 1.9149e-02, 1.0672e-02, 8.3179e-03],
[-5.1511e-03, -2.1840e-02, -1.2597e-02]],
[[-1.1448e-02, -5.5772e-03, -4.4170e-03],
[ 1.3190e-02, 4.8222e-03, 1.1920e-02],
[ 2.1151e-02, -1.6081e-02, -1.6144e-02]],
[[ 1.3584e-02, 8.9827e-03, -2.1343e-02],
[-1.2230e-02, 2.7788e-02, 2.3000e-03],
[-1.7698e-02, -2.2797e-03, -5.8428e-03]],
...,
[[-1.7119e-02, -1.3688e-02, -1.7270e-02],
[ 2.0308e-03, 4.9227e-03, -7.2063e-03],
[ 1.2536e-02, 4.1750e-03, 5.4226e-03]],
[[-9.3082e-03, -7.7180e-03, -1.0361e-03],
[-1.5341e-02, 7.7777e-03, -1.3469e-02],
[-8.7009e-03, -1.2160e-02, -1.3999e-02]],
[[ 3.0552e-02, 1.3847e-02, 1.0074e-02],
[ 3.2691e-02, 2.2141e-02, 2.6602e-02],
[ 2.1128e-02, 1.8867e-02, 2.3278e-02]]],
...,
[[[-5.8947e-03, 2.2211e-02, 8.4547e-03],
[-1.4809e-02, 7.6867e-03, 5.8644e-03],
[-1.1171e-02, 3.2800e-02, 1.4729e-02]],
[[ 1.2799e-02, 3.3404e-03, -8.0116e-03],
[ 4.8239e-04, 2.8425e-03, -2.3057e-03],
[-6.2140e-03, 3.7630e-03, -2.6030e-02]],
[[-7.2594e-03, -2.1482e-02, -2.5453e-02],
[ 3.2283e-02, 2.0096e-03, -7.4505e-03],
[ 1.8465e-02, 2.2960e-04, -8.8664e-04]],
...,
[[-9.5123e-04, 1.4586e-02, 7.2230e-03],
[-5.7989e-03, 5.6743e-03, 1.2585e-02],
[-8.1923e-03, -3.2287e-02, -1.8241e-02]],
[[-1.7783e-02, -6.6726e-03, -4.9502e-03],
[-5.5236e-03, 2.3151e-02, 1.7009e-02],
[ 3.1129e-03, 4.9039e-03, -3.2202e-03]],
[[ 6.8004e-03, 3.6363e-03, -2.0004e-03],
[-1.3302e-02, 8.8023e-03, 9.2807e-04],
[ 4.0554e-03, 8.5429e-04, -6.7752e-04]]],
[[[ 2.5518e-03, 1.4680e-02, 2.6033e-03],
[-1.0130e-02, 9.3898e-03, -5.2180e-03],
[ 5.4886e-03, 2.5363e-03, 1.3153e-03]],
[[-1.0852e-03, -1.4529e-02, -6.6705e-03],
[ 2.0345e-03, -6.7807e-04, -4.9220e-03],
[ 3.5024e-03, 2.1662e-02, 9.4759e-03]],
[[ 2.8206e-04, 1.0020e-02, -2.4434e-02],
[-1.1955e-02, 1.1725e-02, -5.1801e-04],
[-9.4785e-03, 9.1012e-03, -3.6745e-03]],
...,
[[-6.2103e-03, 7.7493e-03, -1.5079e-02],
[ 5.9956e-03, 2.1847e-02, 7.9742e-03],
[ 8.1026e-03, 1.8854e-02, 2.2557e-02]],
[[-1.2351e-02, -8.2933e-03, -8.4467e-03],
[ 7.8022e-03, -4.6217e-03, -2.0065e-04],
[ 1.4313e-03, -7.1164e-03, -4.9417e-03]],
[[-1.5546e-02, -1.5347e-02, -2.1985e-02],
[-4.6449e-03, -4.8014e-03, -6.0732e-03],
[ 7.9861e-03, 1.2879e-02, -6.0797e-03]]],
[[[-3.3188e-02, -1.3365e-02, 2.2784e-02],
[ 1.1430e-02, -4.2369e-03, 3.5036e-03],
[ 3.2804e-02, -1.2503e-02, -1.4610e-02]],
[[-2.0241e-02, -9.5616e-03, -1.7951e-03],
[-5.1452e-04, 2.3131e-03, -4.7360e-03],
[ 4.2102e-03, 3.8810e-03, -3.0991e-03]],
[[-2.3415e-02, -1.2256e-02, -1.6172e-04],
[-3.4541e-02, -1.5332e-02, 5.2646e-03],
[-1.0936e-02, 8.1166e-03, 1.2481e-02]],
...,
[[ 9.7134e-03, 3.2788e-03, 1.1143e-02],
[-8.3905e-03, -1.3490e-02, 9.7034e-03],
[ 9.1477e-03, 1.8615e-03, 2.3738e-02]],
[[-6.7199e-03, 5.7582e-03, -1.5496e-03],
[-9.9920e-03, 1.1892e-02, 6.0360e-03],
[-1.0405e-02, 2.1521e-03, -1.0671e-02]],
[[ 7.0692e-03, -7.7802e-03, -1.0909e-02],
[ 1.5003e-02, -1.6051e-02, -2.7195e-03],
[ 2.0354e-02, 7.5280e-03, 5.8080e-03]]]], device='cuda:0')), ('layer3.4.bn2.weight', tensor([ 0.2008, 0.1521, 0.1259, 0.1868, 0.1921, 0.2461, 0.1508,
0.2064, 0.1900, 0.1617, 0.2026, 0.2085, 0.2002, 0.1725,
0.1931, 0.2046, 0.1522, 0.1449, 0.1675, 0.1770, 0.1903,
0.1664, 0.1221, 0.1943, 0.1635, 0.1830, 0.1265, 0.1798,
0.2328, 0.1720, 0.1510, 0.1793, 0.2229, 0.1740, 0.1833,
0.2031, 0.1971, 0.1704, 0.1564, 0.1749, 0.2048, 0.1699,
0.1695, 0.1954, 0.1749, 0.1589, 0.2077, 0.1882, 0.1825,
0.1804, 0.1969, 0.1586, 0.1490, 0.2586, 0.2114, 0.1672,
0.1810, 0.1974, 0.1204, 0.1673, 0.1668, 0.0994, 0.1337,
0.2054, 0.1776, 0.1594, 0.1904, 0.2033, 0.1793, 0.1668,
0.1542, 0.2019, 0.1837, 0.1943, 0.1338, 0.1116, 0.1929,
0.1416, 0.1832, 0.1873, 0.1796, 0.2255, 0.2024, 0.1473,
0.1653, 0.1743, 0.1756, 0.1637, 0.1556, 0.1973, 0.1373,
0.2583, 0.1816, 0.1601, 0.1976, 0.2355, 0.1635, 0.1721,
0.1581, 0.1209, 0.1857, 0.1934, 0.1604, 0.1342, 0.1788,
0.1542, 0.1111, 0.2232, 0.2222, 0.1729, 0.2066, 0.2073,
0.1874, 0.1871, 0.1798, 0.1855, 0.1945, 0.1648, 0.1276,
0.2362, 0.1991, 0.1864, 0.1613, 0.1551, 0.1677, 0.2030,
0.1779, 0.2047, 0.2305, 0.1639, 0.1974, 0.1774, 0.1617,
0.2104, 0.1987, 0.1616, 0.1128, 0.1850, 0.1840, 0.1340,
0.1449, 0.1235, 0.1881, 0.1881, 0.2245, 0.2144, 0.1606,
0.1581, 0.1136, 0.2017, 0.1562, 0.1934, 0.1712, 0.1811,
0.2129, 0.1608, 0.1805, 0.1978, 0.1429, 0.2109, 0.1914,
0.2316, 0.1665, 0.1306, 0.1375, 0.2008, 0.1785, 0.2253,
0.1964, 0.1796, 0.1488, 0.1971, 0.1903, 0.1901, 0.1814,
0.2048, 0.1558, 0.1998, 0.1498, 0.1917, 0.2003, 0.1697,
0.2062, 0.1638, 0.1716, 0.1496, 0.2007, 0.1685, 0.2033,
0.1396, 0.2072, 0.2134, 0.1713, 0.1970, 0.1853, 0.1743,
0.1968, 0.1930, 0.1577, 0.1861, 0.1883, 0.1416, 0.2113,
0.1806, 0.1282, 0.1883, 0.1662, 0.1733, 0.1812, 0.1679,
0.1543, 0.2277, 0.2178, 0.1982, 0.1462, 0.1987, 0.1771,
0.1761, 0.1810, 0.1655, 0.1544, 0.1721, 0.2057, 0.1552,
0.1237, 0.1702, 0.1612, 0.1703, 0.1313, 0.1579, 0.1547,
0.2015, 0.1264, 0.1486, 0.1426, 0.1809, 0.1765, 0.1920,
0.1950, 0.1873, 0.1197, 0.1519, 0.2036, 0.1775, 0.1829,
0.1793, 0.1870, 0.2008, 0.1836, 0.1925, 0.1903, 0.1819,
0.2041, 0.1113, 0.1589, 0.2006], device='cuda:0')), ('layer3.4.bn2.bias', tensor([-0.1534, 0.0185, 0.0063, -0.1076, -0.1880, -0.1379, -0.0456,
-0.1717, -0.1251, 0.0141, -0.1812, -0.1499, -0.1637, -0.0188,
-0.1679, -0.1444, -0.0716, -0.0381, -0.0721, -0.1751, -0.1522,
-0.0850, 0.0063, -0.0695, -0.0497, -0.1304, 0.0265, -0.0965,
-0.1659, -0.0703, -0.0357, -0.0827, -0.1077, -0.1052, -0.0943,
-0.1322, -0.0909, -0.0434, -0.0348, -0.0832, -0.0884, -0.0801,
-0.0949, -0.1403, -0.0660, -0.0954, -0.1571, -0.1181, -0.1045,
-0.0982, -0.1383, -0.0774, -0.0425, -0.3299, -0.1648, -0.0514,
-0.0858, -0.1613, 0.0429, -0.0169, -0.0669, 0.1415, -0.0169,
-0.1406, -0.1162, -0.0315, -0.0810, -0.1463, -0.1556, -0.0816,
-0.0693, -0.1055, -0.1247, -0.1693, -0.0105, 0.2091, -0.1337,
0.0016, -0.1023, -0.1028, -0.0606, -0.1522, -0.1760, 0.0498,
-0.0562, -0.1344, -0.1116, -0.0575, -0.0552, -0.1528, -0.0358,
-0.1000, -0.1304, -0.0047, -0.1353, -0.1303, -0.0915, -0.0978,
-0.0104, 0.0226, -0.0793, -0.1432, -0.0402, -0.0381, -0.1285,
-0.0798, 0.0532, -0.0730, -0.2105, -0.0254, -0.1120, -0.1362,
-0.1390, -0.0606, -0.1243, -0.1169, -0.1203, -0.1249, 0.0253,
-0.1063, -0.0950, -0.0806, -0.0616, -0.0574, -0.0497, -0.1602,
-0.1168, -0.1691, -0.1608, -0.0488, -0.1329, -0.1264, -0.0928,
-0.0536, -0.1786, -0.0642, 0.0652, -0.0765, -0.0856, -0.0073,
-0.0301, 0.0740, -0.0872, -0.1576, -0.1412, -0.1435, -0.0439,
-0.0509, 0.0339, -0.0990, -0.1011, -0.1460, -0.0592, -0.1022,
-0.1250, -0.0839, -0.1574, -0.1794, -0.0208, -0.0967, -0.0815,
-0.1702, -0.0819, 0.0412, -0.0051, -0.2341, -0.1041, -0.1077,
-0.1686, -0.1207, -0.0321, -0.1361, -0.1332, -0.1180, -0.1132,
-0.1542, -0.0331, -0.1595, -0.0496, -0.1201, -0.1465, -0.0696,
-0.1338, -0.0479, -0.0996, -0.0330, -0.0948, -0.0846, -0.1685,
0.0679, -0.0419, -0.0863, -0.0898, -0.1544, -0.1005, -0.0553,
-0.1426, -0.1411, -0.0633, -0.1525, -0.0995, 0.0178, -0.1673,
-0.0620, 0.0531, -0.1369, -0.0632, -0.1179, -0.1617, -0.1062,
-0.0803, -0.1628, -0.1020, -0.1734, -0.0217, -0.1389, -0.1538,
-0.0584, -0.1068, -0.0700, -0.0484, -0.1114, -0.1421, -0.0315,
0.0264, -0.0764, -0.0019, -0.0895, 0.0459, -0.1228, -0.0460,
-0.1303, 0.0304, -0.0031, -0.0399, -0.1261, -0.0755, -0.0842,
-0.1189, -0.1167, 0.0110, -0.0680, -0.1387, -0.1016, -0.1808,
-0.1114, -0.1655, -0.1242, -0.0829, -0.1266, -0.1015, -0.1227,
-0.1013, 0.0410, -0.0963, -0.1434], device='cuda:0')), ('layer3.4.bn2.running_mean', tensor([-0.1098, 0.0811, -0.0985, -0.0899, -0.0780, -0.1177, -0.0444,
-0.0685, -0.0706, 0.0419, -0.1137, -0.0801, -0.0695, -0.1068,
-0.0585, -0.0873, -0.0636, -0.0364, -0.0473, -0.0260, -0.0414,
-0.0776, -0.1128, -0.1583, -0.0713, -0.0710, 0.0577, -0.0588,
0.0356, -0.0394, -0.0534, -0.0745, -0.1182, -0.0773, -0.0657,
-0.0760, -0.0750, 0.0312, -0.0670, -0.0523, -0.1123, -0.0651,
0.0014, -0.0606, -0.0409, -0.1362, -0.0547, -0.1155, -0.1072,
-0.0582, -0.0188, -0.0987, 0.0410, 0.0304, -0.0460, -0.0603,
-0.0744, -0.1355, -0.0557, -0.1293, -0.0119, -0.0526, -0.1386,
-0.0697, -0.1005, -0.0737, -0.0827, -0.0220, -0.0689, -0.0946,
-0.0314, -0.0869, -0.1064, -0.1163, -0.1108, 0.1168, -0.0311,
0.0548, -0.1086, -0.0688, -0.1294, -0.1270, -0.0695, 0.0693,
0.1132, -0.1091, -0.1273, -0.1078, -0.0479, -0.0562, -0.0056,
-0.0880, -0.0202, -0.1778, -0.0961, -0.0251, 0.1160, -0.1113,
0.0197, 0.0535, -0.1331, -0.0679, -0.0220, -0.0817, -0.0619,
-0.0170, 0.0804, -0.1629, -0.1220, 0.0596, -0.1495, -0.0483,
-0.1196, -0.0787, -0.1021, -0.0701, -0.0368, -0.0730, -0.0567,
-0.1851, -0.1494, -0.0593, -0.0873, -0.0159, -0.0471, -0.1062,
0.2870, -0.1224, -0.1162, -0.0645, -0.0529, -0.0706, -0.1467,
-0.0996, -0.0497, -0.0566, 0.0698, -0.0375, 0.0085, -0.0209,
-0.0446, 0.0323, -0.1066, -0.1009, 0.1744, -0.0764, -0.0585,
0.0313, 0.0009, 0.0419, 0.0844, -0.1134, -0.1141, -0.1217,
-0.1299, -0.1205, -0.0411, -0.0974, -0.0831, -0.0549, -0.0996,
-0.0945, -0.1301, -0.0109, -0.0186, -0.0234, -0.1304, -0.1040,
-0.1129, 0.0810, -0.0067, -0.0919, -0.2165, -0.1210, -0.0986,
-0.0733, -0.0170, -0.0646, -0.0608, -0.0345, -0.1238, -0.1109,
-0.1083, -0.1234, -0.0580, -0.0981, -0.0406, -0.0302, -0.0782,
0.0239, -0.0965, -0.1185, -0.0318, -0.0840, -0.0979, -0.0685,
-0.1081, -0.1545, -0.0757, -0.1397, 0.0165, -0.1100, -0.1235,
-0.0991, -0.0864, -0.0743, -0.0068, -0.0304, -0.1128, -0.0798,
0.0420, -0.0729, 0.0759, -0.1321, -0.0477, -0.0791, -0.1329,
-0.0839, -0.0954, -0.1073, -0.0609, -0.0967, -0.0763, -0.1149,
-0.0275, -0.0839, -0.0268, -0.1182, -0.0723, -0.0895, -0.1041,
-0.0479, 0.0641, -0.0600, -0.0543, -0.0671, -0.0591, -0.0724,
-0.1299, -0.0815, -0.0746, -0.0200, 0.0885, -0.1264, 0.1632,
-0.1117, -0.1337, -0.0837, -0.0852, -0.1253, -0.1640, -0.0560,
0.0398, -0.0048, -0.0457, -0.0639], device='cuda:0')), ('layer3.4.bn2.running_var', tensor(1.00000e-02 *
[ 0.7647, 1.3358, 1.0169, 0.6767, 0.6833, 1.2945, 0.8098,
0.6125, 0.8586, 1.9364, 0.8598, 1.0367, 0.6401, 0.9573,
0.5275, 1.2335, 0.6753, 0.6721, 1.0848, 0.8738, 0.7776,
0.6425, 0.8618, 1.0083, 0.7882, 1.0923, 0.7326, 0.8263,
0.9403, 0.7926, 0.6742, 1.4767, 1.2050, 0.6829, 0.7424,
0.7703, 1.4614, 0.7734, 0.7680, 0.5898, 1.7852, 0.6389,
0.4721, 0.7974, 0.7006, 0.7631, 0.6544, 0.8299, 1.3487,
0.8174, 0.8822, 0.7360, 0.6234, 2.9646, 0.7835, 0.7565,
0.5751, 0.9992, 0.9208, 1.1992, 0.6874, 0.9579, 0.7161,
0.6451, 0.6735, 0.6115, 1.9468, 1.8218, 0.8860, 1.5465,
0.7385, 1.0001, 0.8530, 1.1403, 1.1571, 1.2329, 0.5791,
1.8320, 0.8240, 0.7082, 1.2030, 0.7618, 0.8528, 1.3900,
0.6330, 0.8571, 1.2267, 0.7278, 0.9611, 0.6214, 0.6280,
1.8523, 0.7287, 0.9930, 0.8546, 2.9354, 0.8102, 0.6785,
1.7342, 0.9269, 0.8969, 0.9662, 0.6971, 0.6465, 0.6406,
0.7307, 0.8387, 1.1035, 0.6668, 1.9231, 0.9982, 0.8333,
0.7844, 0.9743, 0.8028, 0.7104, 0.8483, 0.6279, 1.0728,
1.2680, 1.2637, 1.8572, 0.8456, 0.5304, 0.8833, 0.8717,
1.6002, 0.9811, 1.3592, 0.7456, 0.8729, 0.6813, 0.7336,
1.1568, 0.6536, 0.6846, 0.7417, 0.9405, 0.8993, 0.5566,
0.7398, 1.2390, 0.9482, 0.7964, 1.7783, 1.3334, 0.7735,
1.0139, 0.9529, 0.9028, 0.7759, 1.0337, 1.1648, 1.0361,
1.3651, 0.6396, 0.5616, 0.7352, 0.7354, 1.1358, 0.8268,
0.8301, 1.0551, 1.4724, 0.9197, 0.7436, 0.6756, 1.4046,
0.6026, 0.9665, 0.7455, 1.0438, 0.7770, 0.9456, 0.8051,
0.7247, 1.0037, 0.8176, 0.8492, 0.8023, 1.1031, 0.8286,
1.1487, 0.8830, 0.5634, 0.5551, 0.9720, 0.7118, 0.7470,
1.5039, 2.4215, 2.2736, 0.6355, 0.8202, 1.0837, 1.1054,
1.0484, 1.1019, 0.5411, 0.9939, 0.6671, 0.8232, 1.0598,
0.9916, 0.8212, 0.5220, 0.6409, 0.5302, 0.5799, 0.6413,
0.6910, 0.7622, 1.2379, 0.9798, 0.6148, 1.0494, 0.7946,
0.6839, 0.9275, 0.8376, 0.6141, 0.8939, 0.9010, 0.5708,
1.0120, 0.7080, 1.3698, 0.9011, 0.7925, 0.5999, 0.7157,
0.9184, 0.7566, 0.7914, 0.8792, 1.3277, 0.7986, 1.1157,
0.9925, 0.7285, 0.8040, 0.6268, 1.4538, 0.9772, 1.2853,
1.1405, 0.7570, 1.4851, 1.1742, 0.9133, 1.5456, 0.6830,
1.7043, 0.8105, 0.5399, 0.8661], device='cuda:0')), ('layer3.4.conv3.weight', tensor([[[[-5.9331e-04]],
[[ 1.8079e-03]],
[[ 2.0339e-03]],
...,
[[-4.1875e-03]],
[[ 9.9313e-04]],
[[ 1.1720e-02]]],
[[[-2.6283e-02]],
[[-3.2600e-03]],
[[-2.4356e-02]],
...,
[[-1.3290e-02]],
[[-3.0868e-02]],
[[ 3.9922e-03]]],
[[[-1.2431e-02]],
[[ 2.8370e-02]],
[[-1.7048e-03]],
...,
[[ 6.5235e-03]],
[[-6.0346e-03]],
[[-2.8201e-02]]],
...,
[[[-9.0503e-04]],
[[ 1.3985e-02]],
[[ 2.7736e-02]],
...,
[[ 2.9050e-02]],
[[-5.3157e-03]],
[[ 2.8942e-02]]],
[[[ 1.5963e-02]],
[[-1.8018e-02]],
[[ 9.1698e-03]],
...,
[[ 2.8396e-03]],
[[ 6.5481e-03]],
[[ 1.1409e-02]]],
[[[-2.2462e-02]],
[[ 3.6912e-04]],
[[ 1.0604e-02]],
...,
[[ 3.9585e-03]],
[[-6.8926e-03]],
[[-3.9098e-03]]]], device='cuda:0')), ('layer3.4.bn3.weight', tensor([ 0.0009, 0.1826, 0.1513, ..., 0.1095, 0.0555, 0.0358], device='cuda:0')), ('layer3.4.bn3.bias', tensor([-0.0169, -0.2128, -0.1429, ..., -0.0969, -0.0237, -0.0253], device='cuda:0')), ('layer3.4.bn3.running_mean', tensor([ 0.0095, -0.0120, -0.0060, ..., -0.0223, -0.0240, 0.0022], device='cuda:0')), ('layer3.4.bn3.running_var', tensor(1.00000e-03 *
[ 0.1894, 1.2409, 0.9745, ..., 0.7920, 0.5885, 0.3003], device='cuda:0')), ('layer3.5.conv1.weight', tensor([[[[-6.0310e-04]],
[[ 1.7761e-02]],
[[ 4.6879e-02]],
...,
[[ 2.6565e-03]],
[[-3.1731e-02]],
[[-3.1961e-02]]],
[[[ 2.8313e-02]],
[[-2.9960e-02]],
[[ 6.8789e-03]],
...,
[[ 9.8169e-04]],
[[ 1.3060e-02]],
[[ 3.7413e-03]]],
[[[ 2.9517e-02]],
[[-3.3596e-03]],
[[ 1.2223e-02]],
...,
[[-2.2092e-02]],
[[ 1.6818e-02]],
[[ 6.9128e-03]]],
...,
[[[ 2.6911e-03]],
[[-1.8400e-03]],
[[ 1.2463e-02]],
...,
[[ 1.5551e-03]],
[[-5.0797e-03]],
[[ 1.7978e-02]]],
[[[ 1.7029e-02]],
[[-2.0254e-03]],
[[-1.6941e-03]],
...,
[[-1.4131e-02]],
[[ 1.2477e-02]],
[[-2.8510e-02]]],
[[[-2.7551e-02]],
[[ 5.7712e-02]],
[[-1.1912e-02]],
...,
[[ 1.9937e-02]],
[[-1.9331e-02]],
[[-7.3146e-03]]]], device='cuda:0')), ('layer3.5.bn1.weight', tensor([ 0.1948, 0.2103, 0.1212, 0.2151, 0.2013, 0.2338, 0.2944,
0.1917, 0.1829, 0.1640, 0.1770, 0.2103, 0.1740, 0.1739,
0.1975, 0.1661, 0.1908, 0.1852, 0.2386, 0.2910, 0.2078,
0.1500, 0.1946, 0.1704, 0.1506, 0.1702, 0.1703, 0.1529,
0.1867, 0.1885, 0.1863, 0.1446, 0.1716, 0.1494, 0.1802,
0.1402, 0.1357, 0.1655, 0.1934, 0.1497, 0.2556, 0.1460,
0.1724, 0.2584, 0.2165, 0.1947, 0.1798, 0.2354, 0.1708,
0.1231, 0.1407, 0.2105, 0.2117, 0.1465, 0.1701, 0.1808,
0.1243, 0.1922, 0.1521, 0.1621, 0.2064, 0.1426, 0.1856,
0.1543, 0.2007, 0.1708, 0.1919, 0.1408, 0.1831, 0.1552,
0.2117, 0.2639, 0.1480, 0.1831, 0.2081, 0.2090, 0.1943,
0.2180, 0.1667, 0.1511, 0.1736, 0.2208, 0.2237, 0.1889,
0.1799, 0.0897, 0.1993, 0.1611, 0.1971, 0.2159, 0.1882,
0.1870, 0.2000, 0.1844, 0.1307, 0.1710, 0.1956, 0.1528,
0.1681, 0.2044, 0.1870, 0.1935, 0.1741, 0.1656, 0.2113,
0.1651, 0.1835, 0.2180, 0.2107, 0.2069, 0.1844, 0.2376,
0.1798, 0.1882, 0.2032, 0.2480, 0.1715, 0.2192, 0.1886,
0.1920, 0.1843, 0.1610, 0.2126, 0.2105, 0.1596, 0.2083,
0.1651, 0.2035, 0.1743, 0.0949, 0.1796, 0.1529, 0.2042,
0.1798, 0.1606, 0.1607, 0.1746, 0.2506, 0.1223, 0.1925,
0.1924, 0.2121, 0.1935, 0.1857, 0.1669, 0.2394, 0.2276,
0.2252, 0.1508, 0.1717, 0.1808, 0.1765, 0.1416, 0.2039,
0.1914, 0.1832, 0.2311, 0.2254, 0.2168, 0.1956, 0.1924,
0.2333, 0.1844, 0.1900, 0.1785, 0.1613, 0.1865, 0.1906,
0.2555, 0.1293, 0.2190, 0.1616, 0.0984, 0.2218, 0.1602,
0.2044, 0.1798, 0.1376, 0.1238, 0.2010, 0.1835, 0.1742,
0.1905, 0.1863, 0.2193, 0.1538, 0.1164, 0.1184, 0.1739,
0.1561, 0.1640, 0.1818, 0.1947, 0.1763, 0.1744, 0.1801,
0.1632, 0.1439, 0.1758, 0.1852, 0.2062, 0.1236, 0.2004,
0.1913, 0.1859, 0.2046, 0.1666, 0.2145, 0.2149, 0.2011,
0.2430, 0.2123, 0.1989, 0.1607, 0.1499, 0.2037, 0.1780,
0.1772, 0.1461, 0.1630, 0.1615, 0.1403, 0.1770, 0.1623,
0.2794, 0.2106, 0.2188, 0.1711, 0.1859, 0.2209, 0.1645,
0.1755, 0.1464, 0.1461, 0.1962, 0.1848, 0.1843, 0.1541,
0.2093, 0.1778, 0.1801, 0.1604, 0.1644, 0.1697, 0.1581,
0.1750, 0.1816, 0.2000, 0.2390, 0.2000, 0.1911, 0.1910,
0.2021, 0.1601, 0.1581, 0.2218], device='cuda:0')), ('layer3.5.bn1.bias', tensor([-0.1785, -0.2099, 0.0802, -0.1658, -0.1965, -0.2649, -0.3144,
-0.0887, -0.1429, -0.0583, -0.1153, -0.2108, -0.0852, -0.1223,
-0.1581, -0.1357, -0.1542, -0.1896, -0.2577, -0.2571, -0.1749,
0.0310, -0.1794, -0.0743, -0.0657, -0.0592, -0.0611, -0.1266,
-0.1670, -0.1209, -0.1311, -0.0373, -0.0707, -0.0690, -0.0851,
-0.0012, 0.0047, -0.1133, -0.0909, -0.1065, -0.2593, 0.0044,
-0.0931, -0.2399, -0.1464, -0.1072, -0.1299, -0.1336, -0.1267,
0.0212, -0.0075, -0.1517, -0.1415, 0.0184, -0.0890, -0.1225,
0.0779, -0.1476, -0.0708, -0.0566, -0.1248, -0.0549, -0.1264,
-0.0319, -0.1575, -0.0975, -0.1478, -0.0235, -0.1250, -0.0867,
-0.1991, -0.1774, -0.0249, -0.0916, -0.1465, -0.2073, -0.1081,
-0.1618, -0.0178, -0.0159, -0.1207, -0.2073, -0.2235, -0.1227,
-0.1143, 0.1364, -0.1229, -0.0937, -0.1521, -0.2278, -0.1377,
-0.1255, -0.1556, -0.0900, -0.0239, -0.1024, -0.2163, -0.0313,
-0.0678, -0.1471, -0.1437, -0.1314, -0.1286, -0.0424, -0.1573,
-0.0780, -0.0815, -0.1611, -0.1554, -0.1146, -0.0302, -0.1876,
-0.1179, -0.1574, -0.1549, -0.2130, -0.0870, -0.1054, -0.0972,
-0.1181, -0.0949, -0.0361, -0.1453, -0.1136, -0.0825, -0.1286,
-0.1012, -0.1371, -0.1124, 0.1322, -0.0724, -0.0367, -0.1939,
-0.1150, -0.0140, -0.0929, -0.1021, -0.1714, 0.1359, -0.1186,
-0.0458, -0.1254, -0.1112, -0.0745, -0.0571, -0.2073, -0.1167,
-0.1942, -0.0252, -0.1083, -0.1104, -0.1240, 0.0227, -0.2183,
-0.1357, -0.1065, -0.1628, -0.2177, -0.1620, -0.1555, -0.1785,
-0.1714, -0.1276, -0.1518, -0.1752, -0.0056, -0.1766, -0.0879,
-0.2702, 0.0020, -0.2040, -0.0641, 0.0797, -0.1826, -0.0647,
-0.1214, -0.0979, 0.0018, 0.0326, -0.1189, -0.0942, -0.0361,
-0.1732, -0.1679, -0.1091, -0.0713, 0.0521, 0.0196, -0.1274,
-0.0747, -0.1058, -0.0559, -0.1106, -0.1207, -0.1368, -0.0935,
-0.0353, -0.0506, -0.0538, -0.0936, -0.1119, 0.0077, -0.1345,
-0.1528, -0.1444, -0.1633, -0.1036, -0.2081, -0.0842, -0.1674,
-0.1317, -0.1487, -0.1678, -0.0215, -0.0322, -0.1122, -0.1295,
-0.0814, -0.0010, -0.0139, -0.1405, 0.0197, -0.1477, -0.0987,
-0.0793, -0.1658, -0.1859, -0.1489, -0.1157, -0.1785, -0.0796,
-0.1393, -0.0174, -0.0157, -0.1570, -0.1530, -0.1477, -0.0691,
-0.0899, -0.1629, -0.1087, -0.0536, -0.0294, -0.0746, -0.0863,
-0.1033, -0.1157, -0.1933, -0.2118, -0.1378, -0.1393, -0.1381,
-0.1260, 0.0618, -0.0584, -0.1395], device='cuda:0')), ('layer3.5.bn1.running_mean', tensor([-0.0561, -0.0910, 0.0090, -0.1724, 0.0340, -0.1653, -0.1445,
-0.0364, -0.0060, -0.0269, -0.0088, -0.0661, -0.0443, -0.0254,
-0.0340, -0.0898, -0.0639, -0.0621, -0.1637, -0.1667, -0.0996,
-0.0954, -0.0521, -0.0420, -0.0485, -0.0880, 0.0329, -0.0782,
-0.0617, 0.0319, -0.0461, -0.0168, -0.0475, -0.1082, -0.0489,
0.0421, -0.0942, 0.0155, -0.1045, 0.0437, -0.1067, 0.1813,
-0.0296, -0.0771, -0.0775, -0.0853, -0.0635, -0.1703, -0.0537,
-0.0743, -0.0467, -0.0634, -0.1162, -0.0738, -0.0670, -0.0434,
-0.1184, 0.0153, -0.0227, -0.1312, -0.1677, -0.0050, -0.0971,
-0.0377, 0.0096, -0.0153, -0.0960, -0.0501, -0.0938, -0.0028,
-0.0111, -0.1007, -0.0533, -0.0654, -0.1325, -0.0839, 0.0156,
-0.1062, -0.1239, -0.0945, -0.0648, -0.0169, -0.0237, -0.0004,
-0.1111, -0.0635, -0.0751, -0.0181, -0.0568, -0.0959, -0.0150,
0.0174, -0.0562, -0.0900, -0.0019, 0.0159, -0.0634, -0.0541,
-0.0062, -0.0123, -0.0938, -0.0686, -0.1222, -0.0102, -0.0732,
0.0061, -0.1394, -0.0874, -0.0698, -0.0823, 0.0493, -0.0890,
-0.1013, -0.0502, 0.0493, -0.0857, -0.0255, -0.1396, -0.1034,
-0.0540, 0.0490, -0.0179, -0.0476, -0.0606, -0.0532, -0.0219,
-0.0059, -0.0912, -0.1057, -0.0377, -0.0434, -0.0092, -0.0930,
-0.0185, -0.0614, -0.1036, -0.0445, -0.0221, 0.1028, -0.0776,
-0.0691, -0.0438, -0.0418, -0.0001, -0.0399, -0.0702, -0.0035,
-0.1338, -0.0286, 0.0009, -0.0192, -0.0448, -0.0686, -0.1248,
-0.0557, -0.0477, -0.1294, -0.1008, -0.0411, -0.0513, -0.0498,
-0.0523, 0.0264, -0.0350, -0.1138, -0.0721, -0.0138, 0.0164,
-0.1198, -0.0696, -0.0615, 0.0121, -0.0336, -0.1062, -0.1063,
-0.1249, -0.0450, -0.0909, -0.0057, -0.0223, -0.0103, -0.1426,
-0.0174, -0.0626, 0.0097, 0.0220, -0.0541, -0.0316, 0.0050,
-0.0286, 0.0005, -0.1497, -0.0596, -0.0160, -0.0123, -0.0275,
-0.1231, 0.0304, -0.1801, -0.0777, -0.1150, -0.1416, -0.0925,
-0.0531, -0.1011, -0.0481, -0.0291, -0.0801, -0.0621, -0.1151,
-0.1582, -0.0427, -0.0170, -0.1187, 0.0102, -0.1047, -0.1384,
-0.0519, -0.0750, -0.0678, -0.0418, -0.0624, -0.0636, -0.1128,
-0.1435, 0.0018, -0.0593, -0.0551, -0.0595, -0.1248, -0.0409,
0.0240, -0.1110, -0.0220, -0.0058, -0.0322, -0.0578, -0.0681,
-0.0559, -0.0484, -0.0736, 0.0340, -0.1131, -0.0734, -0.0867,
-0.0862, -0.0060, -0.0349, -0.0180, -0.1104, -0.0372, -0.0203,
-0.0437, -0.0268, -0.0009, -0.0375], device='cuda:0')), ('layer3.5.bn1.running_var', tensor(1.00000e-02 *
[ 1.2753, 1.2199, 1.6633, 1.4018, 1.1882, 1.3495, 1.6501,
1.5835, 0.9771, 0.9466, 1.0676, 1.4050, 1.1425, 1.1916,
1.4109, 1.1010, 1.0846, 0.8853, 1.5976, 1.8768, 1.2941,
2.7854, 1.0474, 1.1754, 1.2573, 1.1126, 1.2414, 1.5151,
1.0858, 1.3652, 1.2636, 1.3178, 1.6330, 1.1119, 1.1078,
1.9166, 1.3284, 1.1260, 1.5931, 0.9954, 1.1822, 2.4937,
1.1324, 1.2288, 1.3585, 1.4409, 1.3919, 1.4199, 1.0439,
1.6071, 1.2629, 1.6789, 1.2220, 2.0974, 1.3257, 1.2356,
2.8939, 1.2244, 1.0918, 1.2943, 1.1293, 1.2890, 1.8264,
1.5910, 1.3354, 1.3207, 0.8355, 1.3170, 1.8935, 1.1296,
1.4917, 3.0055, 1.8412, 1.3039, 1.0488, 1.1887, 1.4648,
1.0157, 2.2753, 1.1709, 1.2129, 1.5417, 1.2384, 1.4780,
1.1637, 1.1294, 1.2911, 0.9067, 1.1247, 1.3979, 1.0062,
1.2596, 1.2152, 1.0096, 0.9775, 0.8100, 0.7852, 2.1009,
1.4074, 1.4876, 1.2365, 1.5029, 1.1583, 1.5893, 1.2617,
1.4701, 1.2674, 1.6800, 1.4916, 1.1289, 2.7772, 1.5290,
1.3560, 1.0836, 1.2181, 1.2734, 1.5668, 4.8231, 1.2337,
1.4087, 2.1091, 1.5028, 1.9534, 1.4886, 0.9441, 2.5071,
1.4321, 1.6785, 1.3768, 1.5254, 1.5614, 1.4413, 1.3437,
1.0461, 2.0961, 1.5463, 1.3068, 1.8911, 3.3043, 1.8605,
1.8648, 2.5141, 1.8658, 1.8127, 1.8258, 1.2503, 1.2436,
1.0846, 1.4070, 1.2426, 1.6021, 1.0217, 1.7835, 1.0402,
1.1837, 1.3258, 1.9173, 1.7152, 1.2421, 0.9003, 1.2388,
1.5053, 1.2291, 1.1650, 1.1667, 2.8731, 1.0572, 1.5846,
1.5073, 1.9604, 1.0092, 1.0821, 1.3047, 1.4503, 1.0938,
1.2509, 1.3125, 1.4543, 2.1785, 1.2299, 1.6213, 1.7394,
0.9381, 0.8706, 1.8314, 1.2819, 1.5785, 1.4303, 0.9538,
1.0956, 1.0943, 2.5140, 1.6305, 1.1933, 1.0335, 1.1606,
1.1890, 1.0851, 1.8908, 1.2567, 1.2478, 1.4503, 1.4651,
1.1005, 0.8879, 0.9089, 0.9442, 1.0014, 1.6741, 0.9122,
2.0700, 1.1443, 0.9685, 1.7834, 1.3974, 1.4986, 1.2735,
1.2201, 2.0329, 2.0120, 1.0843, 2.0525, 0.9867, 1.0916,
4.4702, 1.2341, 1.0822, 1.1975, 1.9170, 1.2099, 1.1573,
1.1784, 1.4038, 1.3337, 1.1273, 1.3862, 1.1571, 1.5086,
3.5067, 1.0693, 1.2442, 1.4726, 1.7431, 1.8007, 1.0756,
1.0252, 1.4411, 1.0986, 1.6399, 1.3101, 1.4516, 1.0758,
1.3445, 4.9376, 1.4695, 1.5901], device='cuda:0')), ('layer3.5.conv2.weight', tensor([[[[-1.8232e-02, -3.7203e-02, -2.3618e-02],
[ 3.0867e-03, 2.6153e-02, -6.7986e-03],
[ 9.4285e-04, 2.7014e-04, -7.5859e-03]],
[[ 1.6146e-02, 3.4563e-02, 1.9232e-02],
[ 3.4687e-02, 6.0032e-02, 3.0663e-02],
[-9.6985e-03, 8.1054e-03, -5.2946e-03]],
[[-1.4234e-02, -4.3377e-03, -4.3874e-03],
[-2.9364e-03, -1.6176e-02, 1.6481e-03],
[ 7.3642e-03, 4.9375e-03, 1.0650e-02]],
...,
[[ 2.2973e-03, 1.2850e-02, 7.2108e-03],
[-1.6075e-02, -9.9125e-04, -8.4743e-03],
[-2.4514e-02, -5.8070e-03, 6.4657e-04]],
[[-5.7512e-03, 5.9755e-03, 7.1584e-04],
[-1.9273e-03, -1.6972e-03, 1.2136e-02],
[-5.5451e-03, 4.0342e-03, -7.4075e-04]],
[[-2.2081e-03, -4.3551e-03, 1.6138e-02],
[-1.9957e-04, -5.3369e-03, 8.8725e-03],
[ 1.7258e-03, 2.5699e-03, 2.1026e-03]]],
[[[ 1.2546e-02, -5.8992e-04, 1.1483e-02],
[ 8.9914e-03, -1.7501e-02, 2.2883e-02],
[-1.0734e-03, 2.9461e-03, 1.4120e-02]],
[[-2.1133e-02, -2.2209e-02, -1.1183e-02],
[-1.7590e-02, -1.1415e-02, 8.1762e-03],
[ 1.4293e-02, 1.0924e-02, 1.3077e-02]],
[[ 4.5858e-03, 1.4184e-02, 3.5841e-03],
[ 5.9022e-03, 2.7697e-03, -5.5906e-03],
[ 3.5660e-03, 6.7106e-04, -2.4308e-02]],
...,
[[ 4.1768e-03, -5.8112e-03, -1.4056e-02],
[ 7.0462e-03, -7.4015e-03, -2.3515e-02],
[ 7.2506e-03, -2.1789e-02, -2.8072e-02]],
[[-2.9742e-02, -1.8196e-02, -2.6209e-02],
[-1.0129e-02, 5.7325e-03, -2.9642e-03],
[-1.6088e-02, -1.6614e-02, -1.1019e-02]],
[[-8.2624e-03, -1.6960e-02, -3.3549e-03],
[-1.7180e-03, 6.2526e-03, -3.3032e-03],
[-1.8702e-02, -1.8744e-02, -2.3722e-02]]],
[[[-1.9137e-02, -2.7025e-02, -2.1552e-02],
[-2.2563e-02, -2.3228e-02, -2.6779e-02],
[-8.2968e-03, -6.4610e-03, -1.5783e-02]],
[[ 9.5157e-03, 1.2182e-02, 1.2291e-02],
[-2.8040e-03, -3.4145e-02, -4.8298e-03],
[ 2.2046e-02, 4.7675e-02, 2.1649e-02]],
[[-2.0278e-03, 8.4803e-04, 6.6914e-03],
[ 1.9825e-03, 3.2314e-02, 4.6029e-03],
[-4.1501e-02, -2.1296e-02, -3.4869e-02]],
...,
[[ 4.7098e-04, 8.1674e-03, 2.3782e-02],
[-1.6303e-02, 7.0586e-04, 4.3863e-03],
[-1.2259e-02, -4.9076e-05, -8.3181e-03]],
[[ 1.4525e-02, 2.8155e-02, 1.7697e-02],
[-5.9662e-03, 6.7936e-03, -8.2686e-03],
[ 6.9453e-04, -4.3367e-03, -8.2057e-03]],
[[-1.8842e-02, -2.1245e-02, -2.3877e-02],
[-1.7169e-02, -2.7942e-02, -1.0753e-02],
[-8.7937e-03, -1.1407e-02, 5.4173e-03]]],
...,
[[[-1.9409e-02, -5.9409e-03, 5.9442e-03],
[-1.3004e-02, -3.1706e-02, -5.4351e-03],
[-1.8653e-02, -3.2165e-02, -1.7299e-02]],
[[-2.1062e-02, -1.3807e-02, -6.0481e-03],
[-3.0783e-03, 7.8395e-04, 2.6406e-03],
[ 1.6623e-02, 3.9630e-02, 1.7395e-02]],
[[ 7.0678e-03, -1.0764e-02, -7.9355e-03],
[ 1.7459e-02, -1.3725e-02, 1.7816e-02],
[ 3.9210e-02, 1.7379e-03, 3.2920e-02]],
...,
[[-1.7285e-02, 4.9075e-03, 3.6420e-02],
[-1.3927e-02, -2.3735e-02, 2.0495e-02],
[ 6.6044e-04, -8.5418e-03, 3.0780e-02]],
[[-1.5714e-03, -7.0358e-06, 1.0681e-02],
[ 5.5914e-03, -1.3363e-02, 8.8432e-03],
[ 2.1575e-03, 2.5597e-02, 1.6927e-02]],
[[-2.8245e-02, -2.2754e-02, -2.1844e-02],
[-1.8798e-02, -1.2453e-02, -3.1665e-02],
[-1.6448e-02, -1.8883e-02, -9.7382e-03]]],
[[[ 1.4527e-02, 9.7653e-03, 1.8607e-02],
[ 1.7390e-02, -1.9561e-03, -5.7478e-04],
[ 1.6686e-02, 2.3338e-02, 1.6452e-02]],
[[-4.0944e-04, -1.8717e-02, -1.3497e-02],
[-2.3124e-02, -1.6939e-02, -2.5140e-02],
[-1.3033e-02, -8.9127e-03, -1.2479e-02]],
[[-3.5714e-03, 4.4862e-03, -6.0032e-03],
[-1.0248e-02, 9.9325e-03, 5.7604e-04],
[-1.6890e-02, -1.7979e-02, -1.3478e-02]],
...,
[[-1.4564e-02, -2.6356e-03, -4.5432e-03],
[-1.6651e-02, 4.7062e-03, 4.6228e-03],
[-4.2661e-02, -5.1634e-03, 2.0140e-02]],
[[-2.2336e-02, -6.3959e-03, -1.4273e-02],
[-1.2428e-02, -3.4726e-04, -5.6012e-03],
[-1.1977e-03, 7.1808e-03, 1.0123e-02]],
[[-1.3417e-02, -8.7234e-03, -3.6504e-03],
[-1.3646e-02, 1.1942e-03, -1.1993e-02],
[-1.0109e-02, -2.3825e-03, 4.6304e-05]]],
[[[-1.8207e-02, -1.4984e-02, -5.5921e-03],
[-5.2332e-03, -9.4499e-03, 5.2049e-03],
[-1.6134e-02, -2.5237e-02, 3.4571e-03]],
[[ 6.7473e-04, -1.6203e-02, -7.4866e-03],
[-3.3327e-03, -1.3687e-02, -6.5016e-03],
[-1.1313e-03, -1.4965e-02, -7.7389e-03]],
[[ 6.2745e-03, -2.6325e-03, 9.0637e-03],
[-1.1776e-02, -2.6573e-02, -2.3888e-03],
[ 4.6322e-05, -1.2211e-02, 1.0205e-02]],
...,
[[-2.3483e-02, -1.4903e-02, 1.8685e-02],
[-1.4492e-02, -1.0298e-02, -5.6091e-03],
[-2.9446e-02, -1.0531e-02, 3.6425e-02]],
[[-3.9796e-03, -3.1082e-03, 5.5262e-03],
[ 7.4617e-03, -3.9119e-03, 2.0406e-02],
[ 2.9055e-03, 1.1029e-02, 3.7552e-03]],
[[-2.2511e-02, -2.0920e-02, -7.0648e-03],
[-7.2441e-03, 3.7443e-03, 6.4946e-03],
[-2.3425e-02, -2.0029e-02, -1.6306e-02]]]], device='cuda:0')), ('layer3.5.bn2.weight', tensor([ 0.1799, 0.1605, 0.1783, 0.2058, 0.1696, 0.1842, 0.1896,
0.1804, 0.1934, 0.1578, 0.2080, 0.1694, 0.1916, 0.1874,
0.1915, 0.1741, 0.1861, 0.1983, 0.1571, 0.1738, 0.1989,
0.1918, 0.2507, 0.1985, 0.2673, 0.2015, 0.1865, 0.1705,
0.1490, 0.1483, 0.2136, 0.1862, 0.2224, 0.1555, 0.2088,
0.2230, 0.1968, 0.2047, 0.1893, 0.2631, 0.2627, 0.1734,
0.1242, 0.2153, 0.1686, 0.1458, 0.1808, 0.1640, 0.2165,
0.2037, 0.1304, 0.2112, 0.2024, 0.2059, 0.1898, 0.2017,
0.2588, 0.2188, 0.2253, 0.1665, 0.1293, 0.1915, 0.1621,
0.1938, 0.2011, 0.2077, 0.1975, 0.1982, 0.1697, 0.1678,
0.2291, 0.1479, 0.1789, 0.1988, 0.1555, 0.2127, 0.1735,
0.1964, 0.1736, 0.2242, 0.1795, 0.1817, 0.1719, 0.1855,
0.1770, 0.1576, 0.1541, 0.1984, 0.2344, 0.2019, 0.2195,
0.1836, 0.1987, 0.1910, 0.2012, 0.1697, 0.1727, 0.1793,
0.2071, 0.1914, 0.2501, 0.1884, 0.1887, 0.1951, 0.2129,
0.2244, 0.1791, 0.1999, 0.2044, 0.1941, 0.1926, 0.1959,
0.1806, 0.1630, 0.2154, 0.1732, 0.2303, 0.2440, 0.1516,
0.1596, 0.1556, 0.1869, 0.2629, 0.2113, 0.1619, 0.1587,
0.1845, 0.1732, 0.1779, 0.1950, 0.2191, 0.1859, 0.1843,
0.1634, 0.1895, 0.2074, 0.1900, 0.1853, 0.2092, 0.1943,
0.1716, 0.2214, 0.1908, 0.1776, 0.1515, 0.2557, 0.1530,
0.1435, 0.1909, 0.1779, 0.1919, 0.1913, 0.1885, 0.2086,
0.1791, 0.1657, 0.2026, 0.1309, 0.1823, 0.1708, 0.1615,
0.2154, 0.1943, 0.1827, 0.1934, 0.1686, 0.1740, 0.1768,
0.1884, 0.1395, 0.1694, 0.1249, 0.1984, 0.2558, 0.1818,
0.1458, 0.2029, 0.2144, 0.1988, 0.1616, 0.1437, 0.2796,
0.2465, 0.1809, 0.2152, 0.2142, 0.2673, 0.1915, 0.1817,
0.2041, 0.1874, 0.1996, 0.1443, 0.1502, 0.1848, 0.1803,
0.2029, 0.2030, 0.2153, 0.4324, 0.1349, 0.2063, 0.2048,
0.2260, 0.1979, 0.1902, 0.2182, 0.2398, 0.1996, 0.1711,
0.1908, 0.2316, 0.1845, 0.2276, 0.1680, 0.2185, 0.2005,
0.1392, 0.1781, 0.1988, 0.2038, 0.1930, 0.1915, 0.2092,
0.1931, 0.1841, 0.1957, 0.1809, 0.2423, 0.1645, 0.1514,
0.2635, 0.2045, 0.1309, 0.2280, 0.5254, 0.1907, 0.1380,
0.1668, 0.1868, 0.1888, 0.1843, 0.1799, 0.1814, 0.2375,
0.1715, 0.2021, 0.1671, 0.1971, 0.2151, 0.1966, 0.1898,
0.1715, 0.2070, 0.1768, 0.2672], device='cuda:0')), ('layer3.5.bn2.bias', tensor([-0.0950, -0.0429, -0.0648, -0.1112, -0.1368, -0.0984, -0.1285,
-0.0326, -0.1890, -0.0512, -0.1933, -0.0564, -0.1662, -0.1181,
-0.1060, -0.0185, -0.0933, -0.0582, -0.0318, -0.1021, -0.1560,
-0.0770, -0.1286, -0.0659, -0.1850, -0.0679, -0.0762, -0.0659,
-0.0736, -0.0134, -0.1070, -0.0684, -0.0896, -0.0029, -0.1198,
-0.0119, -0.1063, -0.0686, -0.0612, -0.1319, -0.1781, -0.0765,
0.1662, -0.1244, -0.0856, -0.0270, -0.1248, -0.0530, -0.2007,
-0.0635, 0.1194, -0.1497, -0.0671, -0.1742, -0.1035, -0.1167,
-0.1681, -0.1027, -0.1438, -0.0405, 0.0730, -0.0770, -0.0578,
-0.0834, -0.2014, -0.1179, -0.0407, -0.0703, 0.0021, -0.0678,
-0.1495, -0.0342, -0.0983, -0.1059, -0.0215, -0.0554, -0.0783,
-0.0742, -0.0601, -0.0918, -0.0507, -0.1486, -0.0749, -0.0901,
-0.0635, -0.0740, -0.0870, -0.0976, -0.2599, -0.0890, -0.0899,
-0.1529, -0.0424, -0.1403, -0.1159, -0.0523, -0.0813, -0.1198,
-0.1376, -0.0852, -0.0677, -0.1657, -0.1012, -0.1307, -0.0798,
-0.2427, -0.0658, -0.0631, -0.1021, -0.1274, -0.0958, -0.0602,
-0.1203, -0.0622, -0.0727, -0.1410, -0.0888, -0.0385, 0.0194,
-0.0481, -0.0208, -0.0979, -0.1236, -0.1103, -0.0105, -0.0171,
-0.0954, -0.1581, -0.1261, -0.0830, -0.1141, -0.0670, -0.0860,
-0.0589, -0.1430, -0.1792, -0.1794, -0.1547, -0.1523, -0.0766,
-0.0451, -0.2206, -0.0898, -0.0892, 0.0237, -0.0974, -0.0226,
0.0019, -0.1125, -0.0307, -0.0494, -0.0957, -0.1024, -0.0940,
-0.0093, -0.0868, -0.1182, 0.0850, -0.0677, -0.0792, -0.0409,
-0.0909, -0.1315, -0.0968, -0.0748, -0.0583, -0.1061, -0.1413,
-0.1347, -0.0359, -0.0432, 0.0624, -0.0906, -0.0642, -0.0414,
0.0111, -0.1396, -0.0983, -0.1266, -0.0408, -0.0016, -0.2592,
-0.0847, -0.0986, -0.0259, -0.0874, -0.1198, -0.1240, -0.0968,
-0.1471, -0.0967, -0.1273, 0.0119, -0.0201, -0.0717, -0.0867,
-0.0837, -0.1474, -0.1657, -0.1853, 0.0859, -0.1579, -0.0908,
-0.1340, -0.1316, -0.0701, -0.1214, -0.1153, -0.1148, -0.0618,
-0.0698, -0.1352, -0.0835, -0.1696, -0.0338, -0.1497, -0.0935,
0.0735, -0.0710, -0.1276, -0.0829, -0.1407, -0.1013, -0.1726,
-0.0950, -0.0895, -0.1470, -0.0467, -0.0776, -0.0146, 0.0136,
-0.2587, -0.0910, 0.1044, -0.1138, -0.3993, -0.0544, 0.0161,
-0.0538, -0.1187, -0.0699, -0.0692, -0.0611, -0.0670, -0.2354,
-0.0573, -0.1812, -0.0524, -0.1779, -0.1660, -0.1221, -0.0657,
-0.0900, -0.0665, -0.0885, -0.0772], device='cuda:0')), ('layer3.5.bn2.running_mean', tensor([-0.1188, -0.0727, -0.0450, 0.0143, -0.0618, -0.0296, -0.0832,
-0.0444, -0.0585, -0.0179, -0.0935, -0.0736, -0.1212, -0.0728,
-0.1015, -0.1049, 0.2363, -0.0908, -0.0767, 0.0607, -0.0915,
-0.1603, 0.0820, -0.0303, 0.0108, -0.1121, -0.0982, -0.2030,
-0.0450, -0.0795, -0.0663, -0.0896, -0.1077, -0.0964, -0.0748,
-0.0260, -0.1325, -0.1086, -0.0812, -0.0828, -0.0862, 0.0054,
-0.0195, -0.1562, -0.0601, -0.0453, 0.0492, -0.0866, -0.1233,
-0.0763, 0.0393, -0.0736, -0.0453, -0.0383, -0.0249, -0.0472,
-0.1015, -0.1271, -0.0681, -0.0588, -0.0052, -0.0473, -0.0682,
-0.0500, -0.1601, -0.0707, -0.1616, -0.1066, -0.0599, -0.0142,
0.0262, -0.1179, -0.0512, -0.0448, -0.0930, -0.1219, 0.0029,
-0.0299, -0.0763, -0.0625, -0.0329, -0.0735, -0.0639, -0.0884,
-0.0983, -0.0385, -0.0152, -0.0930, -0.0888, 0.0210, -0.1262,
0.0285, -0.0691, -0.0473, 0.1040, -0.0736, -0.0463, -0.0942,
-0.0436, -0.1023, -0.1008, -0.1103, -0.0570, -0.0310, -0.0721,
0.1923, -0.0770, -0.0293, -0.0369, -0.0360, -0.1011, -0.0983,
-0.0896, 0.0095, -0.0562, -0.0706, -0.0448, -0.0837, -0.0138,
-0.0597, -0.1216, -0.0876, -0.1280, -0.0747, -0.0194, -0.0913,
-0.0601, -0.0711, -0.0793, -0.0916, -0.0094, -0.0549, -0.0521,
-0.1187, -0.1527, -0.0942, -0.1125, -0.0438, -0.0877, -0.0881,
-0.1107, -0.0036, 0.0652, 0.0775, -0.0686, -0.1025, 0.0021,
-0.0722, -0.0672, -0.1300, -0.0351, 0.0011, -0.0518, -0.0932,
-0.0372, -0.0685, -0.0777, -0.1159, -0.1492, -0.0638, -0.1009,
-0.0972, -0.0903, -0.0077, 0.0028, -0.0543, -0.0326, -0.0597,
-0.0145, -0.0557, -0.1068, 0.0185, -0.0817, -0.1195, -0.1410,
-0.0499, 0.0108, -0.1456, -0.0502, -0.0476, -0.0966, -0.1244,
-0.0698, 0.0570, -0.0469, -0.0990, 0.0408, -0.0897, -0.0538,
0.0163, -0.0790, -0.0240, -0.0584, -0.0466, -0.0007, -0.0688,
-0.0561, -0.0394, -0.0270, -0.1036, 0.0379, -0.0498, -0.0654,
0.1477, -0.0877, -0.0755, -0.0081, -0.0726, 0.0191, -0.0798,
-0.1332, -0.0001, -0.1254, 0.0305, -0.0623, -0.0486, -0.0293,
-0.1392, -0.1407, -0.0128, -0.1316, 0.0974, 0.0060, -0.0652,
-0.0809, -0.0868, -0.0733, -0.0248, -0.0931, -0.1936, -0.0159,
-0.2014, -0.1139, 0.0392, -0.0578, -0.2584, -0.0996, -0.0213,
-0.0757, -0.0461, -0.0125, -0.0073, -0.0533, -0.0689, -0.0660,
-0.0491, -0.1731, -0.0864, 0.1501, -0.0277, -0.0283, -0.1244,
-0.0185, -0.0747, -0.0981, -0.1239], device='cuda:0')), ('layer3.5.bn2.running_var', tensor([ 0.0078, 0.0086, 0.0098, 0.0119, 0.0057, 0.0103, 0.0083,
0.0192, 0.0078, 0.0115, 0.0110, 0.0117, 0.0090, 0.0084,
0.0135, 0.0156, 0.0203, 0.0146, 0.0067, 0.0160, 0.0074,
0.0134, 0.0277, 0.0145, 0.0120, 0.0186, 0.0131, 0.0121,
0.0075, 0.0077, 0.0098, 0.0110, 0.0150, 0.0118, 0.0100,
0.0286, 0.0111, 0.0178, 0.0112, 0.0175, 0.0122, 0.0096,
0.0126, 0.0096, 0.0078, 0.0125, 0.0067, 0.0093, 0.0142,
0.0256, 0.0115, 0.0118, 0.0129, 0.0106, 0.0112, 0.0103,
0.0104, 0.0188, 0.0102, 0.0117, 0.0171, 0.0088, 0.0112,
0.0089, 0.0083, 0.0100, 0.0158, 0.0159, 0.0193, 0.0075,
0.0119, 0.0082, 0.0079, 0.0085, 0.0126, 0.0161, 0.0110,
0.0108, 0.0128, 0.0102, 0.0088, 0.0084, 0.0089, 0.0101,
0.0106, 0.0083, 0.0120, 0.0114, 0.0106, 0.0109, 0.0194,
0.0066, 0.0107, 0.0067, 0.0097, 0.0098, 0.0079, 0.0092,
0.0082, 0.0124, 0.0142, 0.0083, 0.0166, 0.0095, 0.0159,
0.0200, 0.0171, 0.0134, 0.0087, 0.0087, 0.0168, 0.0284,
0.0070, 0.0096, 0.0314, 0.0063, 0.0128, 0.1029, 0.0116,
0.0112, 0.0104, 0.0099, 0.0718, 0.0158, 0.0103, 0.0130,
0.0086, 0.0057, 0.0074, 0.0160, 0.0157, 0.0121, 0.0110,
0.0082, 0.0078, 0.0102, 0.0152, 0.0090, 0.0083, 0.0117,
0.0085, 0.0090, 0.0081, 0.0103, 0.0192, 0.0198, 0.0100,
0.0113, 0.0120, 0.0267, 0.0126, 0.0109, 0.0073, 0.0154,
0.0131, 0.0114, 0.0111, 0.0115, 0.0132, 0.0070, 0.0116,
0.0153, 0.0106, 0.0094, 0.0121, 0.0096, 0.0085, 0.0125,
0.0075, 0.0113, 0.0114, 0.0173, 0.0127, 0.0178, 0.0122,
0.0107, 0.0094, 0.0161, 0.0071, 0.0092, 0.0134, 0.0119,
0.0166, 0.0122, 0.0149, 0.0108, 0.0153, 0.0090, 0.0070,
0.0093, 0.0160, 0.0101, 0.0123, 0.0151, 0.0141, 0.0092,
0.0211, 0.0078, 0.0157, 0.0283, 0.0197, 0.0074, 0.0228,
0.0187, 0.0138, 0.0079, 0.0106, 0.0151, 0.0077, 0.0147,
0.0189, 0.0095, 0.0093, 0.0160, 0.0138, 0.0080, 0.0071,
0.0140, 0.0094, 0.0082, 0.0126, 0.0102, 0.0090, 0.0106,
0.0096, 0.0096, 0.0093, 0.0115, 0.0252, 0.0182, 0.0080,
0.0176, 0.0079, 0.0188, 0.0086, 0.1011, 0.0115, 0.0110,
0.0076, 0.0088, 0.0147, 0.0143, 0.0110, 0.0125, 0.0084,
0.0148, 0.0097, 0.0083, 0.0136, 0.0093, 0.0098, 0.0153,
0.0098, 0.0134, 0.0104, 0.0289], device='cuda:0')), ('layer3.5.conv3.weight', tensor([[[[-9.7815e-04]],
[[ 2.4410e-03]],
[[ 6.4362e-04]],
...,
[[ 4.5518e-04]],
[[ 3.1839e-03]],
[[ 1.5780e-04]]],
[[[ 1.2811e-02]],
[[ 1.2113e-02]],
[[ 5.8799e-02]],
...,
[[-2.8383e-02]],
[[-2.0166e-03]],
[[ 5.0809e-03]]],
[[[-4.1260e-02]],
[[-6.2723e-03]],
[[-2.2622e-02]],
...,
[[-1.7216e-03]],
[[ 1.5788e-02]],
[[-3.2657e-02]]],
...,
[[[-3.1535e-02]],
[[ 5.1557e-03]],
[[-1.2088e-02]],
...,
[[ 3.0828e-02]],
[[ 1.3775e-02]],
[[-9.8853e-04]]],
[[[-1.1204e-02]],
[[-7.0668e-03]],
[[ 6.0029e-04]],
...,
[[ 1.0027e-02]],
[[-4.5193e-03]],
[[-2.3091e-02]]],
[[[ 1.1157e-02]],
[[-1.1145e-02]],
[[ 2.1084e-03]],
...,
[[-7.7713e-04]],
[[-1.5691e-02]],
[[-1.2437e-02]]]], device='cuda:0')), ('layer3.5.bn3.weight', tensor([ 0.0096, 0.2409, 0.1293, ..., 0.1258, 0.1065, 0.0620], device='cuda:0')), ('layer3.5.bn3.bias', tensor([-0.0098, -0.2781, -0.1677, ..., -0.1630, -0.0894, -0.0560], device='cuda:0')), ('layer3.5.bn3.running_mean', tensor([-0.0099, -0.0643, -0.0472, ..., -0.0093, -0.0511, -0.0186], device='cuda:0')), ('layer3.5.bn3.running_var', tensor(1.00000e-02 *
[ 0.0368, 0.1920, 0.0873, ..., 0.0947, 0.1198, 0.0872], device='cuda:0')), ('layer4.0.conv1.weight', tensor([[[[-1.2395e-02]],
[[ 8.1686e-03]],
[[ 9.8950e-03]],
...,
[[-1.6895e-02]],
[[-5.7022e-03]],
[[ 1.4820e-02]]],
[[[ 4.2635e-02]],
[[ 3.0915e-02]],
[[ 1.4444e-02]],
...,
[[-1.1674e-02]],
[[-5.1237e-03]],
[[ 1.5901e-02]]],
[[[-6.5722e-03]],
[[ 8.2411e-03]],
[[-8.9546e-03]],
...,
[[-2.0367e-02]],
[[ 2.4588e-02]],
[[-1.0744e-02]]],
...,
[[[-1.2986e-02]],
[[-9.0921e-03]],
[[ 4.8936e-02]],
...,
[[-9.2243e-03]],
[[ 1.4984e-02]],
[[ 1.4262e-02]]],
[[[-1.1871e-02]],
[[-2.1683e-03]],
[[ 3.2217e-02]],
...,
[[-1.2867e-02]],
[[ 2.3403e-02]],
[[-1.0939e-02]]],
[[[ 2.3710e-02]],
[[ 3.5328e-02]],
[[-1.3471e-02]],
...,
[[-3.0454e-02]],
[[-6.8836e-03]],
[[-2.9400e-02]]]], device='cuda:0')), ('layer4.0.bn1.weight', tensor([ 0.2743, 0.2033, 0.1934, 0.2054, 0.2730, 0.1637, 0.1773,
0.2263, 0.2446, 0.2194, 0.1984, 0.2645, 0.1942, 0.2152,
0.2243, 0.2375, 0.2741, 0.3050, 0.2097, 0.1821, 0.1994,
0.2179, 0.1714, 0.1766, 0.2234, 0.2086, 0.2545, 0.2658,
0.2333, 0.2316, 0.2091, 0.1868, 0.2292, 0.1874, 0.1989,
0.1636, 0.2628, 0.2418, 0.2327, 0.2817, 0.1933, 0.2856,
0.2123, 0.2094, 0.2937, 0.2543, 0.2078, 0.2731, 0.2279,
0.1040, 0.2676, 0.2257, 0.1750, 0.2330, 0.2074, 0.2665,
0.2020, 0.1493, 0.2105, 0.2821, 0.1869, 0.1877, 0.2775,
0.2114, 0.2289, 0.1380, 0.2131, 0.1572, 0.2439, 0.2202,
0.2432, 0.1965, 0.2304, 0.1702, 0.2066, 0.2074, 0.1941,
0.1847, 0.2464, 0.2256, 0.2534, 0.2314, 0.2236, 0.1665,
0.2321, 0.2707, 0.1803, 0.2441, 0.1637, 0.2487, 0.2329,
0.1887, 0.2016, 0.1998, 0.1916, 0.2339, 0.2155, 0.2484,
0.2770, 0.2198, 0.2432, 0.2388, 0.2399, 0.2188, 0.2397,
0.2077, 0.2195, 0.2253, 0.2653, 0.2211, 0.2071, 0.2763,
0.1920, 0.2062, 0.2042, 0.2318, 0.2136, 0.2418, 0.2175,
0.2190, 0.2505, 0.1866, 0.2402, 0.2089, 0.2795, 0.2066,
0.2652, 0.2397, 0.2551, 0.2175, 0.2158, 0.2247, 0.1744,
0.2133, 0.2170, 0.2167, 0.2297, 0.1910, 0.2122, 0.2430,
0.2523, 0.2116, 0.3101, 0.2272, 0.2821, 0.1862, 0.2493,
0.1801, 0.2510, 0.2076, 0.2410, 0.1394, 0.2072, 0.2620,
0.1948, 0.2736, 0.2122, 0.1633, 0.1914, 0.2363, 0.2798,
0.3250, 0.2370, 0.2265, 0.2068, 0.2363, 0.2294, 0.2010,
0.1936, 0.1737, 0.2272, 0.2468, 0.2156, 0.2339, 0.2071,
0.2272, 0.2330, 0.2285, 0.2477, 0.2800, 0.2488, 0.2020,
0.1800, 0.2565, 0.1904, 0.2071, 0.2589, 0.2227, 0.2243,
0.1735, 0.2423, 0.2233, 0.2145, 0.1907, 0.2071, 0.1505,
0.2385, 0.1958, 0.2423, 0.2462, 0.1999, 0.2135, 0.2224,
0.1959, 0.2057, 0.2664, 0.1954, 0.1696, 0.2264, 0.1993,
0.2110, 0.1563, 0.2112, 0.2096, 0.2325, 0.2356, 0.1898,
0.2311, 0.2199, 0.2145, 0.2406, 0.2268, 0.2320, 0.2349,
0.2407, 0.2112, 0.2222, 0.2377, 0.2704, 0.2515, 0.2186,
0.2405, 0.2607, 0.2643, 0.2706, 0.2479, 0.1752, 0.2715,
0.2743, 0.2349, 0.1082, 0.2481, 0.2236, 0.2090, 0.2014,
0.2336, 0.1879, 0.2104, 0.2248, 0.1675, 0.2392, 0.2667,
0.2188, 0.2537, 0.2155, 0.2467, 0.1779, 0.2556, 0.2551,
0.1618, 0.1927, 0.2504, 0.1830, 0.2161, 0.2064, 0.2246,
0.2076, 0.1838, 0.2211, 0.1577, 0.2080, 0.2301, 0.2466,
0.2395, 0.2435, 0.2187, 0.2552, 0.1666, 0.1804, 0.2211,
0.1722, 0.2085, 0.2525, 0.2126, 0.2607, 0.2532, 0.2194,
0.2235, 0.2831, 0.2539, 0.1823, 0.2820, 0.2301, 0.2155,
0.1995, 0.1713, 0.2244, 0.1935, 0.1782, 0.1804, 0.1974,
0.2271, 0.2471, 0.2042, 0.1876, 0.2371, 0.2351, 0.1968,
0.2368, 0.2348, 0.2591, 0.2223, 0.1652, 0.2333, 0.2090,
0.2389, 0.2490, 0.2026, 0.2635, 0.2212, 0.2465, 0.2117,
0.1935, 0.1959, 0.2241, 0.1989, 0.2171, 0.1734, 0.2320,
0.2097, 0.1975, 0.2356, 0.2080, 0.2304, 0.1825, 0.2275,
0.2226, 0.1620, 0.2153, 0.2192, 0.2519, 0.1944, 0.2297,
0.2114, 0.2783, 0.1876, 0.2176, 0.2201, 0.2214, 0.2410,
0.1510, 0.2026, 0.2574, 0.2453, 0.2202, 0.2024, 0.2381,
0.1746, 0.2161, 0.2208, 0.2344, 0.2650, 0.1762, 0.1718,
0.1563, 0.1836, 0.3137, 0.2428, 0.2154, 0.2110, 0.3457,
0.2255, 0.2410, 0.2103, 0.2017, 0.2116, 0.2063, 0.2466,
0.1807, 0.2172, 0.2589, 0.1808, 0.2264, 0.2271, 0.2291,
0.2319, 0.1892, 0.2505, 0.2088, 0.2516, 0.2002, 0.2253,
0.2068, 0.2441, 0.2792, 0.1761, 0.1756, 0.1954, 0.2263,
0.2487, 0.2225, 0.2415, 0.3116, 0.2633, 0.2499, 0.2196,
0.2487, 0.2013, 0.2942, 0.2429, 0.2699, 0.2142, 0.2269,
0.2443, 0.2492, 0.1924, 0.1707, 0.2303, 0.2020, 0.2311,
0.2185, 0.2356, 0.2084, 0.1492, 0.2562, 0.2605, 0.2376,
0.2436, 0.2221, 0.2361, 0.2808, 0.2327, 0.2220, 0.2651,
0.2444, 0.2292, 0.2403, 0.2462, 0.0901, 0.1919, 0.2162,
0.2503, 0.2106, 0.2813, 0.2022, 0.1972, 0.2603, 0.2169,
0.2037, 0.2349, 0.2019, 0.2246, 0.1824, 0.2175, 0.2266,
0.2077, 0.2009, 0.2205, 0.1851, 0.1968, 0.2851, 0.1930,
0.2329, 0.2078, 0.2859, 0.2296, 0.2683, 0.1762, 0.2357,
0.2203, 0.2517, 0.2407, 0.2617, 0.1753, 0.1976, 0.2921,
0.2063, 0.2274, 0.1988, 0.2040, 0.2562, 0.2292, 0.2388,
0.2264, 0.2422, 0.2067, 0.3174, 0.2180, 0.2481, 0.2036,
0.1823, 0.1888, 0.2148, 0.1934, 0.2436, 0.2196, 0.2145,
0.1851, 0.2710, 0.1948, 0.1465, 0.2494, 0.1950, 0.2533,
0.2222, 0.2185, 0.2150, 0.2092, 0.2192, 0.2124, 0.2241,
0.2608], device='cuda:0')), ('layer4.0.bn1.bias', tensor([-0.2175, -0.1799, -0.1100, -0.1448, -0.2565, 0.0492, -0.1362,
-0.1713, -0.2351, -0.1041, -0.2077, -0.1958, -0.1803, -0.2104,
-0.2300, -0.1518, -0.2346, -0.2020, -0.2052, -0.0566, -0.1536,
-0.1688, -0.1069, -0.0934, -0.1327, -0.1054, -0.2204, -0.2439,
-0.1228, -0.2453, -0.0927, -0.0740, -0.2245, -0.1156, -0.1716,
-0.1531, -0.1929, -0.1668, -0.1726, -0.1973, -0.1692, -0.1978,
-0.1371, -0.1495, -0.2370, -0.0678, -0.1081, -0.2250, -0.2562,
0.0881, -0.2044, -0.1220, -0.1325, -0.2394, -0.2502, -0.2933,
-0.1789, -0.0857, -0.0777, -0.2873, -0.0991, -0.1661, -0.1882,
-0.1210, -0.1992, -0.1329, -0.1250, -0.0891, -0.2507, -0.1280,
-0.2486, -0.1084, -0.2467, -0.1884, -0.1866, -0.0843, -0.1766,
-0.1606, -0.2554, -0.2136, -0.2223, -0.1978, -0.1158, -0.1188,
-0.1019, -0.1916, -0.0978, -0.1443, -0.0204, -0.2506, -0.2222,
-0.1718, -0.1852, -0.2127, -0.1483, -0.1942, -0.1323, -0.2508,
-0.1629, -0.1066, -0.2699, -0.1526, -0.2800, -0.1130, -0.1015,
-0.2379, -0.1203, -0.1143, -0.2438, -0.2542, -0.1498, -0.2129,
-0.1255, -0.1401, -0.1829, -0.2148, -0.1425, -0.1696, -0.2388,
-0.1738, -0.2457, -0.1164, -0.1230, -0.1068, -0.1772, -0.1987,
-0.3131, -0.1771, -0.1981, -0.1634, -0.2014, -0.1552, -0.1778,
-0.1982, -0.1955, -0.1238, -0.1581, -0.1376, -0.2230, -0.1832,
-0.1514, -0.1347, -0.2579, -0.2358, -0.1123, -0.1921, -0.1213,
-0.2301, -0.2434, -0.2112, -0.1978, -0.0382, -0.2039, -0.0801,
-0.0522, -0.1426, -0.1337, -0.1813, -0.1908, -0.2357, -0.2066,
-0.3014, -0.1814, -0.2796, -0.1295, -0.1445, -0.1720, -0.2024,
-0.2271, -0.1469, -0.2160, -0.2605, -0.1853, -0.2004, -0.1946,
-0.2256, -0.1330, -0.1881, -0.2145, -0.1467, -0.2147, -0.2173,
-0.1638, -0.1645, -0.2412, -0.2380, -0.2369, -0.1201, -0.1737,
-0.1947, -0.2248, -0.2078, -0.1531, -0.2107, -0.2120, -0.1528,
-0.2134, -0.1736, -0.2488, -0.1752, -0.1445, -0.1865, -0.1190,
-0.2022, -0.1366, -0.2623, -0.1333, 0.1102, -0.1365, -0.1397,
-0.2247, -0.1975, -0.0741, -0.1760, -0.1098, -0.1142, -0.1122,
-0.2348, -0.2491, -0.1476, -0.1853, -0.1992, -0.0928, -0.2074,
-0.2273, -0.1146, -0.1794, -0.1899, -0.2414, -0.2695, -0.1660,
-0.2049, -0.2626, 0.0170, -0.1558, -0.1534, -0.0682, -0.2339,
-0.1621, -0.0868, -0.0009, -0.1981, -0.2039, -0.1744, -0.1966,
-0.1256, -0.1532, -0.1582, -0.2419, -0.1034, -0.1754, -0.3058,
-0.2084, -0.2561, -0.1169, -0.1729, -0.1543, -0.1533, -0.1378,
-0.1516, -0.1534, -0.2340, -0.1233, -0.0575, -0.1849, -0.1185,
-0.1718, -0.1530, -0.2071, -0.1567, -0.1873, -0.1241, -0.1158,
-0.2133, -0.2723, -0.0861, -0.1730, -0.1222, -0.1051, -0.1767,
-0.1645, -0.1620, -0.2219, -0.0702, -0.2662, -0.1309, -0.2535,
-0.2318, -0.0976, -0.1478, -0.0542, -0.2169, -0.2793, -0.3006,
-0.1995, -0.0909, -0.1615, -0.1909, -0.0893, -0.1548, -0.0915,
-0.1922, -0.1872, -0.2035, -0.2421, -0.1927, -0.2361, -0.0879,
-0.1535, -0.1416, -0.2241, -0.2073, -0.1433, -0.2264, -0.1961,
-0.1621, -0.2543, -0.2622, -0.2884, -0.2121, -0.2139, -0.1828,
-0.0882, -0.1388, -0.1568, -0.1791, -0.2085, -0.1248, -0.1674,
-0.1549, -0.2171, -0.1388, -0.1234, -0.1261, -0.1753, -0.1634,
-0.2291, -0.0834, -0.1215, -0.0830, -0.3382, -0.2124, -0.2218,
-0.1493, -0.2163, -0.1303, -0.1403, -0.1272, -0.2103, -0.2194,
-0.1042, -0.1952, -0.2575, -0.1614, -0.2015, -0.2818, -0.1492,
-0.1467, -0.1913, -0.1754, -0.2600, -0.1565, -0.1710, -0.1034,
-0.1076, -0.1432, -0.2016, -0.2785, -0.2496, -0.1170, -0.3327,
-0.1343, -0.1531, -0.1055, -0.1559, -0.1722, -0.1906, -0.1888,
-0.0853, -0.1788, -0.2599, -0.0932, -0.2099, -0.1537, -0.2074,
-0.2283, -0.1066, -0.1559, -0.1285, -0.1730, -0.1308, -0.2720,
-0.2018, -0.2023, -0.2041, -0.0971, -0.1356, -0.2794, -0.1635,
-0.2273, -0.1362, -0.3073, -0.2162, -0.2648, -0.1899, -0.3040,
-0.2439, -0.1564, -0.2525, -0.1693, -0.2470, -0.1699, -0.0972,
-0.1665, -0.1809, -0.1186, -0.1976, -0.2621, -0.1708, -0.1551,
-0.2387, -0.2400, -0.1290, 0.1461, -0.2014, -0.2014, -0.2107,
-0.1418, -0.0967, -0.3211, -0.2771, -0.2649, -0.1488, -0.2005,
-0.1182, -0.2908, -0.1524, -0.2102, 0.1331, -0.1809, -0.2389,
-0.1671, -0.1830, -0.1796, -0.2224, -0.2330, -0.2046, -0.1534,
-0.2536, -0.2405, -0.1471, -0.1390, -0.1410, -0.1806, -0.2740,
-0.1542, -0.1148, -0.1059, -0.1150, -0.1049, -0.1884, -0.2190,
-0.2587, -0.1619, -0.2345, -0.0919, -0.1784, -0.2129, -0.2114,
-0.2235, -0.1418, -0.1427, -0.1163, -0.1768, -0.2016, -0.2028,
-0.2096, -0.1989, -0.2729, -0.1899, -0.2448, -0.1911, -0.1738,
-0.2169, -0.1609, -0.1990, -0.2140, -0.1964, -0.2217, -0.2203,
-0.1598, -0.1437, -0.1008, -0.1497, -0.1960, -0.0950, -0.2124,
-0.1454, -0.1416, -0.0996, -0.1401, -0.3130, -0.1790, -0.1652,
-0.2414, -0.0983, -0.1872, -0.0862, -0.1388, -0.1491, -0.2915,
-0.1916], device='cuda:0')), ('layer4.0.bn1.running_mean', tensor([-0.0812, -0.0476, -0.0685, -0.0392, -0.0879, 0.0749, -0.0952,
-0.0131, -0.1293, -0.0271, -0.0560, -0.0858, -0.0256, -0.0830,
-0.0821, -0.0692, -0.0540, -0.1586, -0.0398, -0.0574, -0.0812,
-0.1505, -0.0959, -0.0591, -0.0701, -0.0490, -0.0391, -0.0410,
-0.0267, -0.1101, -0.0800, 0.0014, -0.0553, -0.0757, 0.0099,
-0.0470, -0.1188, -0.0594, -0.1254, -0.0660, -0.1127, -0.0666,
-0.0708, -0.0861, -0.0925, -0.0326, -0.0765, -0.0459, -0.0841,
-0.0155, -0.0373, -0.1073, -0.0417, -0.0434, -0.1154, -0.1298,
-0.0822, -0.0561, -0.0198, -0.1350, -0.0635, -0.1900, -0.0657,
-0.1017, -0.1496, 0.0088, -0.0978, -0.0414, -0.0650, -0.0444,
-0.0513, -0.0408, -0.0799, -0.1031, -0.0769, 0.0019, -0.0535,
-0.0787, -0.0456, -0.1187, -0.0929, -0.2025, -0.0158, -0.0944,
-0.0111, -0.1060, -0.0331, -0.0619, -0.0846, -0.1498, -0.0764,
-0.0975, 0.0008, -0.1088, -0.0495, -0.0564, -0.0955, -0.1055,
-0.0951, -0.0622, -0.0908, -0.0646, -0.0835, -0.1006, -0.0820,
-0.0778, -0.0564, -0.1040, -0.0940, -0.0501, -0.1371, -0.0998,
-0.0541, -0.1130, -0.0482, -0.0847, -0.0583, -0.1275, -0.1059,
-0.0561, -0.0790, -0.0585, -0.0951, -0.0387, -0.0533, -0.0441,
-0.0943, -0.1286, -0.0798, -0.1764, -0.0697, -0.0433, -0.1061,
-0.0872, -0.0613, -0.0552, -0.0298, -0.0143, -0.0375, -0.1110,
-0.1218, -0.1273, -0.1871, -0.0454, -0.0947, -0.1199, -0.1250,
-0.1039, -0.0988, -0.0152, -0.0783, -0.0005, -0.0458, -0.0942,
-0.0467, -0.1387, -0.0964, -0.0112, -0.1472, -0.0503, -0.0368,
-0.0928, -0.0703, -0.1323, -0.0440, -0.0871, -0.0341, -0.1012,
-0.0859, -0.1108, -0.0841, -0.1134, -0.0308, -0.0979, -0.0817,
-0.0450, -0.0850, -0.0337, -0.0778, -0.0519, -0.1097, -0.0627,
-0.0496, -0.0870, -0.0440, -0.0268, -0.0676, -0.0384, -0.1006,
-0.0710, -0.1459, -0.0765, -0.0841, -0.0697, -0.0470, 0.0052,
-0.0775, -0.0293, -0.0954, -0.1106, -0.0760, -0.0493, -0.0391,
-0.0293, -0.0865, -0.0384, -0.0402, 0.0163, -0.1626, -0.0801,
-0.0371, -0.1024, -0.0865, -0.0663, -0.0415, -0.0613, -0.0520,
0.0097, -0.1390, -0.0619, -0.0563, -0.0791, -0.1096, -0.0450,
-0.0620, -0.0064, -0.0727, -0.0712, -0.1396, -0.0942, -0.1096,
-0.0463, -0.0633, 0.1191, -0.0189, -0.1061, -0.0281, -0.0726,
-0.0841, -0.0987, -0.0792, -0.0382, -0.1308, -0.0943, -0.0354,
-0.0998, -0.0862, -0.1294, -0.1868, -0.0041, -0.1077, -0.0857,
-0.0477, -0.1428, -0.0726, -0.1260, -0.0452, -0.1139, -0.1081,
-0.0405, -0.1460, -0.0963, -0.0775, -0.0873, -0.0519, -0.0626,
-0.0561, -0.0539, -0.1086, -0.0125, -0.0643, -0.0751, -0.0667,
-0.0833, -0.0812, -0.0663, -0.0581, -0.0475, -0.0494, -0.0907,
-0.0825, -0.0871, -0.0870, 0.0094, -0.0571, -0.0763, -0.0993,
-0.1011, -0.1036, -0.0638, -0.0613, -0.0785, -0.2049, -0.0859,
-0.0722, -0.0611, -0.1532, -0.0370, -0.0888, -0.0191, -0.0493,
-0.0522, -0.1108, -0.0216, -0.1226, -0.0702, -0.0957, -0.0948,
-0.0624, -0.0761, -0.0676, -0.0283, -0.0426, -0.1052, -0.0744,
-0.1057, -0.0274, -0.1470, -0.1029, -0.0638, -0.0694, -0.0800,
0.0070, -0.0525, -0.0474, -0.0115, -0.0513, -0.0340, -0.1054,
-0.0524, -0.0129, -0.0268, -0.1543, -0.0288, -0.0525, -0.0580,
-0.0686, -0.0565, -0.1327, -0.0530, -0.1184, -0.0355, -0.0832,
-0.1245, -0.0968, -0.1413, -0.0453, -0.1065, -0.0892, -0.1152,
-0.0324, -0.0711, -0.0736, -0.0742, -0.0865, -0.0358, -0.0569,
-0.0530, 0.0034, -0.0624, -0.1117, -0.0910, -0.0906, -0.0989,
-0.0730, -0.0730, -0.1652, -0.0973, -0.0240, -0.0467, -0.0865,
-0.0137, -0.0836, -0.0752, -0.1154, -0.0988, 0.0062, -0.0057,
-0.0748, -0.0840, -0.0478, -0.0383, -0.1740, -0.0036, -0.0620,
-0.0934, -0.0652, -0.0735, -0.0352, -0.0427, -0.0454, -0.0911,
-0.0738, -0.1254, -0.0388, -0.0583, -0.0844, -0.0625, -0.1014,
-0.0657, -0.0797, -0.0459, -0.0502, -0.0661, -0.1051, -0.1029,
-0.1375, -0.1487, -0.1146, -0.1343, -0.1341, -0.0234, 0.0151,
-0.0914, -0.0870, -0.0780, -0.0780, -0.0924, -0.0481, -0.0554,
-0.1025, -0.0467, -0.0269, -0.0144, -0.0554, -0.1481, -0.1306,
-0.0670, -0.0309, -0.1264, -0.0452, -0.1065, -0.0700, -0.1103,
-0.0414, -0.1035, -0.1069, -0.0709, -0.0626, -0.1327, -0.0105,
-0.0163, -0.1067, -0.1305, -0.1786, -0.0977, -0.0579, -0.1381,
-0.0834, -0.0539, -0.0341, -0.0886, -0.0290, 0.0010, -0.1001,
-0.0991, -0.0704, -0.0608, -0.0880, -0.0521, -0.0886, -0.1157,
-0.0649, -0.1203, -0.0563, -0.0931, -0.0794, -0.1132, -0.0814,
-0.0397, -0.0367, -0.0644, -0.0324, -0.0956, -0.0713, -0.0880,
-0.0646, -0.0514, -0.0700, -0.0728, -0.1070, -0.0884, -0.1378,
-0.0907, -0.0840, -0.0659, -0.1956, -0.0566, -0.1031, -0.1508,
-0.0608, -0.0889, -0.0239, -0.1012, -0.0678, -0.0166, -0.0620,
-0.1525, -0.0289, 0.0119, -0.0352, -0.0693, -0.0590, -0.0638,
-0.0125, -0.0040, -0.0760, -0.0376, -0.0115, -0.0852, -0.1041,
-0.0601], device='cuda:0')), ('layer4.0.bn1.running_var', tensor(1.00000e-02 *
[ 1.3726, 1.1629, 2.0442, 1.1953, 1.3950, 1.5547, 1.5067,
1.0455, 1.6629, 1.4880, 1.3005, 1.3644, 1.2959, 1.3635,
1.1588, 1.2148, 1.3475, 2.2851, 1.6421, 1.4626, 1.1320,
1.8274, 1.1831, 1.2274, 1.1178, 1.0935, 1.1499, 1.5161,
1.5867, 1.0685, 1.5491, 1.2533, 1.1333, 1.5374, 1.5731,
1.1999, 1.8335, 1.3444, 1.9029, 1.1797, 1.6215, 1.3534,
1.2448, 1.5733, 1.7243, 2.2063, 1.7989, 1.4695, 1.1507,
2.2705, 1.2058, 1.7062, 1.7173, 0.9905, 1.2301, 1.8217,
1.4541, 1.4477, 1.2452, 1.5913, 1.4986, 2.1706, 1.1947,
1.5611, 2.1145, 1.2970, 1.7438, 2.3111, 1.3062, 1.4982,
1.4682, 0.9973, 1.1249, 1.4357, 1.8018, 1.2368, 1.0970,
1.4880, 1.2056, 1.3874, 1.3157, 1.6746, 1.4935, 1.5951,
1.8232, 1.4246, 1.0927, 1.3470, 1.4202, 1.7410, 1.1122,
1.2712, 1.2868, 1.0887, 1.0468, 1.0047, 1.4809, 1.7436,
2.6515, 1.4088, 1.1027, 2.0858, 2.3701, 1.3469, 1.6626,
1.3919, 1.5220, 1.9242, 1.5747, 1.2413, 1.3789, 1.4718,
1.0759, 1.4179, 1.1272, 1.5851, 1.5216, 1.8353, 1.2933,
1.2696, 1.2689, 1.6968, 1.3515, 1.5035, 1.3544, 1.1723,
1.1582, 1.6982, 0.9593, 1.5520, 1.3229, 1.6712, 1.2022,
1.3669, 1.3092, 1.2329, 1.3224, 1.5710, 1.2934, 1.2935,
1.7427, 1.3371, 2.2271, 1.3859, 2.2936, 1.1395, 1.4063,
1.2354, 1.0694, 1.1808, 1.5073, 1.0753, 1.2655, 2.7393,
1.8965, 1.4599, 1.3639, 1.2422, 1.4943, 1.8333, 1.4157,
1.6201, 1.5057, 1.6389, 1.5697, 1.7164, 1.3969, 1.2031,
1.0839, 1.6334, 1.4160, 1.4192, 1.2111, 1.5966, 1.0311,
1.4552, 1.5666, 1.1213, 1.5457, 2.3607, 1.6271, 1.2514,
0.9735, 1.3574, 1.7412, 1.3227, 1.0900, 1.4758, 1.3529,
1.3304, 2.2171, 1.1354, 2.8129, 1.0632, 1.1875, 1.1989,
1.2202, 1.6151, 1.7714, 1.2967, 1.2259, 1.1220, 1.5064,
1.4315, 1.1922, 1.2511, 1.3667, 2.1947, 1.7716, 1.3734,
1.0865, 1.4924, 2.0351, 0.9577, 1.3386, 1.5082, 1.2170,
1.0899, 1.8038, 1.3049, 1.5734, 0.9777, 1.5674, 1.0812,
1.4954, 1.4715, 1.2939, 1.0595, 1.5258, 1.1898, 2.0008,
1.3235, 1.4457, 3.3456, 1.3179, 1.5777, 1.2666, 1.5446,
1.7291, 1.6248, 1.0322, 1.2368, 1.4527, 1.5198, 1.1644,
2.2527, 1.4088, 1.4557, 1.6602, 0.9886, 2.0638, 1.2268,
1.3053, 1.6149, 1.2461, 1.4653, 1.4155, 1.7982, 2.1566,
1.3461, 1.4710, 1.4915, 1.1561, 1.5462, 1.2931, 1.4835,
1.2039, 0.9945, 1.2240, 1.1020, 1.7717, 1.2795, 1.9010,
1.1665, 1.2440, 1.4613, 1.1485, 2.2138, 1.2905, 1.0889,
1.7377, 1.5230, 1.2989, 1.6544, 1.4048, 2.5700, 1.3912,
1.2913, 2.0093, 1.2885, 1.6274, 1.5786, 1.2948, 0.9812,
1.1625, 1.4127, 1.5101, 1.3108, 1.2743, 1.2188, 1.3332,
1.3312, 1.5212, 1.1391, 1.4409, 1.5720, 1.4982, 1.3470,
1.4496, 1.3467, 1.7106, 1.4655, 0.7867, 2.6512, 1.6048,
1.4561, 1.5525, 1.6024, 1.4597, 1.1281, 1.5194, 1.0595,
1.5476, 1.7254, 1.6304, 1.1905, 1.1405, 1.0114, 1.3538,
1.3773, 1.2049, 1.8635, 2.1279, 1.1150, 1.6624, 1.6157,
1.1840, 1.1455, 1.4957, 1.3887, 1.1214, 1.0494, 1.4940,
1.7786, 1.5287, 1.1165, 1.4249, 1.4676, 1.9644, 1.2831,
1.3390, 1.6829, 1.2951, 1.1406, 1.0347, 1.0043, 1.3373,
1.4633, 1.3343, 1.3212, 1.4751, 1.6475, 1.1513, 1.0577,
1.4894, 1.5620, 2.1912, 1.1501, 1.0111, 1.1869, 1.9788,
1.2291, 1.7193, 1.3030, 1.9124, 1.3691, 1.4482, 1.7079,
1.6975, 1.5739, 1.2928, 1.4985, 1.4841, 1.7601, 1.4779,
1.5103, 2.0667, 1.3592, 1.2780, 1.0169, 1.2187, 1.2125,
1.3664, 1.7342, 2.0068, 1.4672, 1.6352, 1.8146, 1.4868,
1.1948, 1.5091, 1.1019, 1.5380, 1.4073, 1.2439, 1.3336,
1.3342, 1.7720, 1.1729, 1.5859, 1.4214, 1.3444, 1.3872,
1.3925, 1.8021, 1.0786, 1.4228, 1.3781, 1.0560, 1.6580,
1.5827, 1.4081, 1.2366, 2.2449, 0.8969, 1.1956, 1.5247,
1.5478, 1.7235, 1.1266, 1.3185, 1.2479, 1.8588, 1.2753,
1.6109, 1.3522, 1.2075, 1.4003, 2.3162, 1.6756, 1.6720,
1.3653, 1.2905, 1.5428, 2.0539, 1.1312, 1.5947, 1.8491,
1.0350, 1.3335, 1.6799, 1.6609, 1.2136, 1.4618, 1.8875,
1.3184, 1.4198, 1.4893, 1.1131, 1.3443, 2.3740, 1.8762,
1.6770, 1.5720, 1.0728, 1.7467, 1.3896, 1.1706, 1.2617,
1.4095, 1.3597, 1.5718, 1.4371, 1.3788, 1.2423, 1.4289,
1.3570, 1.4353, 1.0773, 1.1013, 1.6611, 1.5855, 1.9961,
1.4844, 1.3633, 1.2687, 2.1601, 1.4903, 1.3847, 1.2555,
1.0499, 1.2460, 1.0416, 1.3607, 1.6337, 1.4831, 1.4511,
2.1949, 1.8152, 1.9207, 1.2708, 1.4795, 1.1816, 1.5473,
1.3930, 1.3497, 1.2171, 1.5060, 1.2856, 1.2317, 1.6907,
1.4090], device='cuda:0')), ('layer4.0.conv2.weight', tensor([[[[-2.7038e-03, -9.6471e-03, 8.6150e-03],
[-3.5489e-03, -2.3885e-03, -1.4506e-04],
[-1.2197e-02, -1.1853e-02, -1.5689e-02]],
[[ 9.6099e-03, 1.2867e-02, 2.3959e-03],
[ 1.4147e-02, 2.6262e-03, 1.4046e-02],
[ 9.0645e-03, 9.0932e-03, 1.8738e-02]],
[[ 1.5832e-02, 1.0862e-02, 1.6110e-02],
[ 7.0888e-03, 2.0931e-03, 1.2161e-02],
[ 3.1605e-03, 3.3018e-03, 6.4014e-03]],
...,
[[-6.7167e-03, -1.1520e-02, -2.6927e-03],
[-1.0045e-05, -9.7624e-03, -1.3708e-02],
[-6.8258e-03, -1.9444e-02, -1.5843e-02]],
[[-1.6252e-03, 1.1240e-02, 6.3921e-03],
[ 2.2122e-03, 2.2450e-02, 1.4773e-02],
[ 2.3398e-02, 2.0348e-02, 1.9558e-02]],
[[-5.0104e-03, -1.9092e-02, -1.2678e-02],
[ 3.1653e-03, -3.6351e-03, 1.2458e-03],
[-5.4769e-04, 1.6287e-03, 3.9111e-03]]],
[[[-8.1116e-03, -6.7215e-03, -1.3508e-02],
[-8.9186e-04, -4.4791e-03, -4.1201e-03],
[-6.2726e-04, -1.3723e-04, -2.1817e-03]],
[[ 5.6819e-03, -1.8985e-02, -1.2669e-02],
[-7.2007e-04, -6.8813e-03, -1.3810e-02],
[-9.2162e-04, -1.7418e-03, -2.7791e-03]],
[[-7.1193e-03, -5.6796e-03, -4.7309e-03],
[-1.2745e-02, -9.5858e-03, -6.5466e-03],
[-1.6635e-03, 3.4890e-03, -1.3560e-02]],
...,
[[ 1.6020e-03, -1.0735e-02, -8.3338e-03],
[-9.6833e-04, -2.6690e-03, -2.3840e-03],
[ 1.2332e-03, 2.9175e-03, 8.9609e-03]],
[[ 6.4942e-03, 1.0978e-02, 1.2764e-02],
[ 6.7255e-03, 2.1111e-02, 9.5538e-03],
[ 1.2166e-02, 2.6638e-02, 2.1545e-02]],
[[ 1.0668e-02, 1.7984e-02, 1.8792e-02],
[ 1.3521e-02, 2.2373e-03, 2.2643e-02],
[ 1.2784e-02, 1.9381e-02, 4.0592e-03]]],
[[[-5.9661e-04, -2.8229e-03, -7.7746e-03],
[-5.5197e-03, -2.5168e-03, -1.1096e-02],
[-5.5650e-04, -2.3068e-04, -1.1738e-02]],
[[-5.5543e-03, 7.6155e-03, 1.8776e-03],
[ 9.2012e-03, 9.1710e-03, 7.9225e-03],
[ 1.6556e-03, 9.9295e-03, 3.0431e-03]],
[[ 1.2069e-02, 3.2276e-03, 1.0645e-02],
[-7.6239e-03, 4.9690e-03, 3.4100e-03],
[ 1.8734e-02, 1.1239e-02, 1.4936e-02]],
...,
[[-1.3944e-02, -7.5021e-03, -2.7554e-03],
[-1.3870e-02, -2.8909e-03, -9.2927e-03],
[-1.3162e-02, -4.1935e-03, -4.4376e-03]],
[[ 4.6537e-03, -6.7037e-04, 1.5026e-03],
[ 5.2301e-03, -3.7320e-03, 8.8545e-03],
[ 9.7061e-04, 8.5330e-03, 7.4374e-03]],
[[ 9.4894e-04, -1.0470e-02, -3.1285e-03],
[-4.2011e-03, -1.0243e-02, -1.4096e-03],
[-8.3806e-03, -2.0252e-02, -3.5286e-03]]],
...,
[[[ 5.3583e-03, 1.0416e-02, 7.2539e-03],
[ 1.9610e-03, -3.3853e-03, -1.2607e-02],
[-2.2333e-03, 3.0116e-03, -9.1997e-03]],
[[-5.1024e-03, -9.5475e-03, -5.3135e-03],
[-1.5920e-02, -2.0028e-02, -2.8013e-02],
[-1.3500e-02, -1.3823e-02, -1.1464e-02]],
[[ 1.1544e-02, 1.4393e-02, 1.1279e-02],
[ 1.1365e-02, 1.1506e-02, 2.7646e-02],
[ 8.5635e-03, 1.5422e-02, 1.3201e-02]],
...,
[[ 7.5883e-04, 1.0777e-02, -2.8765e-04],
[ 9.9223e-03, 1.6462e-02, 1.3143e-02],
[ 1.5402e-02, 8.0672e-03, 1.2939e-02]],
[[ 5.1292e-03, 9.9754e-04, 3.0944e-04],
[-3.2543e-03, -6.4767e-03, 5.8551e-04],
[-4.9410e-03, -1.1929e-03, -4.6806e-03]],
[[ 2.1055e-03, 1.2636e-02, 1.2070e-02],
[ 6.3100e-03, -2.4499e-03, 4.2961e-03],
[ 6.0020e-03, -1.0730e-03, -1.2693e-02]]],
[[[-1.0348e-02, 5.4107e-03, -7.4006e-03],
[-2.0322e-02, -1.9909e-03, 7.6916e-04],
[-1.0732e-02, -3.0568e-03, -1.1182e-02]],
[[-9.3652e-03, -5.4102e-03, -6.7198e-03],
[-1.1117e-02, -9.6830e-03, -1.2045e-02],
[-2.6053e-02, -1.1185e-02, -1.1533e-02]],
[[-7.6776e-03, -3.8431e-03, -4.9616e-03],
[-4.1145e-03, -6.0760e-03, 1.0061e-02],
[-1.4590e-02, -3.9651e-03, 3.3865e-03]],
...,
[[-5.7134e-03, -2.1272e-03, 2.0195e-04],
[-3.5164e-03, 1.0868e-02, 8.0739e-03],
[ 6.1376e-04, 5.8339e-03, 1.1564e-02]],
[[ 1.4493e-02, 3.3504e-02, 2.8916e-02],
[ 1.5184e-02, 3.7341e-03, 2.3456e-02],
[ 1.5684e-02, 2.5286e-02, 3.1715e-02]],
[[-1.0366e-02, -2.2251e-03, 4.4390e-03],
[ 7.3125e-03, 1.3951e-02, 1.9130e-02],
[ 8.8381e-03, -3.7710e-03, 6.8614e-03]]],
[[[-1.1821e-02, -9.9529e-03, -9.8249e-03],
[-6.3562e-03, -1.7930e-02, -1.4820e-02],
[-1.4924e-02, -1.6608e-02, -1.1495e-02]],
[[ 4.4000e-03, 9.8955e-03, 6.1966e-03],
[ 3.5933e-03, 1.3729e-02, 7.3008e-03],
[-5.5458e-03, -3.6055e-03, 6.3775e-03]],
[[-1.1361e-02, -2.0547e-02, -1.4878e-02],
[-9.5037e-03, -1.7843e-02, -1.3277e-02],
[-1.1011e-02, -2.4447e-02, -2.5935e-02]],
...,
[[-1.0171e-02, -8.8960e-03, -1.1316e-02],
[-2.1769e-02, -1.8169e-02, -1.8610e-02],
[-1.4144e-02, -8.1158e-03, -1.5657e-02]],
[[ 1.1591e-02, -2.2063e-03, 7.6175e-03],
[-4.2510e-03, -6.1025e-03, -4.3307e-03],
[ 4.1824e-03, -6.4818e-03, 8.7410e-03]],
[[ 5.3481e-03, 1.0640e-02, 7.2333e-03],
[ 1.0817e-02, -1.9440e-03, 5.4910e-03],
[ 1.9874e-03, -1.1021e-02, -4.0092e-03]]]], device='cuda:0')), ('layer4.0.bn2.weight', tensor([ 0.2164, 0.2073, 0.2007, 0.2360, 0.1912, 0.2223, 0.2223,
0.1922, 0.2082, 0.1989, 0.2278, 0.1981, 0.2363, 0.2150,
0.1925, 0.2349, 0.1974, 0.1843, 0.1278, 0.1822, 0.1828,
0.2131, 0.1834, 0.1877, 0.1743, 0.2030, 0.1720, 0.2248,
0.1891, 0.1953, 0.1851, 0.1675, 0.1882, 0.1728, 0.2456,
0.2076, 0.1744, 0.1846, 0.2230, 0.1664, 0.1645, 0.1481,
0.1867, 0.1998, 0.1616, 0.1598, 0.2261, 0.1773, 0.2624,
0.2223, 0.1751, 0.1899, 0.1454, 0.2058, 0.2316, 0.2128,
0.2070, 0.2037, 0.2330, 0.2170, 0.1880, 0.1878, 0.1794,
0.2324, 0.2106, 0.1693, 0.1920, 0.1808, 0.2217, 0.1851,
0.1585, 0.2050, 0.2070, 0.2042, 0.1669, 0.1625, 0.1797,
0.2019, 0.1641, 0.1923, 0.2118, 0.1995, 0.1949, 0.2136,
0.2127, 0.1672, 0.2501, 0.1896, 0.1836, 0.1792, 0.2235,
0.1801, 0.2223, 0.2044, 0.2838, 0.2102, 0.1539, 0.2389,
0.1550, 0.1518, 0.2254, 0.2083, 0.2120, 0.2503, 0.2032,
0.1911, 0.2263, 0.1840, 0.2149, 0.1914, 0.1932, 0.1452,
0.2009, 0.2005, 0.1511, 0.1718, 0.1732, 0.1793, 0.1578,
0.2141, 0.1731, 0.1919, 0.2093, 0.1797, 0.2170, 0.1479,
0.2192, 0.1932, 0.2408, 0.1514, 0.2043, 0.2067, 0.1884,
0.1882, 0.1813, 0.2194, 0.1534, 0.2094, 0.2512, 0.2439,
0.1966, 0.2228, 0.1652, 0.1640, 0.1761, 0.1667, 0.1407,
0.2024, 0.1926, 0.2047, 0.1833, 0.1849, 0.2037, 0.2367,
0.1860, 0.2355, 0.1798, 0.1696, 0.1729, 0.2050, 0.1679,
0.1774, 0.2057, 0.1922, 0.1723, 0.2048, 0.1898, 0.1503,
0.1799, 0.1688, 0.2074, 0.2121, 0.2338, 0.1652, 0.2176,
0.1852, 0.1760, 0.1634, 0.1908, 0.1965, 0.1658, 0.2177,
0.1937, 0.1672, 0.2000, 0.2020, 0.1886, 0.1873, 0.1601,
0.2048, 0.1765, 0.1815, 0.1613, 0.1674, 0.1852, 0.2184,
0.1648, 0.1677, 0.1967, 0.1914, 0.2216, 0.1663, 0.1830,
0.1636, 0.1748, 0.2183, 0.2006, 0.2598, 0.2104, 0.1811,
0.2232, 0.2023, 0.1380, 0.1807, 0.1839, 0.1960, 0.2178,
0.2067, 0.2245, 0.1523, 0.2219, 0.2094, 0.1907, 0.1564,
0.2376, 0.1814, 0.1565, 0.1732, 0.2081, 0.1773, 0.1922,
0.1654, 0.1890, 0.2205, 0.2472, 0.1870, 0.1942, 0.2039,
0.2017, 0.2127, 0.2036, 0.2226, 0.2198, 0.1806, 0.1775,
0.2008, 0.1954, 0.2305, 0.1734, 0.2126, 0.1651, 0.1717,
0.2350, 0.1965, 0.1977, 0.2238, 0.2501, 0.1963, 0.2039,
0.2562, 0.1764, 0.1792, 0.2081, 0.2016, 0.1990, 0.1764,
0.1397, 0.1999, 0.1788, 0.2033, 0.1606, 0.1695, 0.1612,
0.2239, 0.1632, 0.2117, 0.2033, 0.2071, 0.2293, 0.1626,
0.2134, 0.2379, 0.1684, 0.2292, 0.2115, 0.1739, 0.1941,
0.1939, 0.2174, 0.1442, 0.1673, 0.2137, 0.1669, 0.1934,
0.1988, 0.1961, 0.1783, 0.1908, 0.1660, 0.1856, 0.1950,
0.2227, 0.1884, 0.1679, 0.2212, 0.1815, 0.1957, 0.1940,
0.1963, 0.2160, 0.2240, 0.1865, 0.2052, 0.1847, 0.1727,
0.1759, 0.1960, 0.1380, 0.2533, 0.1896, 0.1706, 0.1579,
0.1891, 0.2036, 0.1955, 0.1904, 0.2703, 0.2337, 0.2217,
0.2438, 0.1883, 0.1991, 0.1958, 0.2060, 0.1764, 0.2159,
0.2136, 0.2360, 0.2164, 0.1727, 0.2643, 0.2018, 0.2030,
0.1871, 0.1762, 0.1966, 0.1472, 0.2218, 0.1531, 0.1865,
0.2614, 0.1882, 0.2162, 0.2110, 0.1349, 0.1869, 0.2086,
0.1976, 0.1992, 0.1774, 0.1688, 0.2028, 0.1707, 0.1682,
0.1414, 0.1871, 0.1851, 0.1939, 0.2233, 0.1926, 0.2037,
0.2050, 0.1837, 0.1750, 0.1982, 0.1933, 0.1721, 0.1646,
0.2129, 0.1411, 0.1933, 0.1828, 0.1787, 0.2292, 0.2304,
0.1588, 0.2216, 0.1996, 0.1958, 0.2197, 0.2504, 0.2000,
0.1918, 0.1886, 0.2170, 0.2020, 0.1863, 0.2402, 0.1553,
0.1723, 0.2612, 0.2046, 0.2164, 0.1802, 0.1564, 0.2090,
0.1604, 0.1897, 0.1774, 0.2297, 0.2078, 0.2128, 0.2013,
0.2169, 0.2287, 0.2390, 0.1932, 0.1816, 0.1553, 0.1815,
0.1818, 0.2215, 0.2006, 0.1963, 0.2298, 0.1763, 0.2521,
0.1975, 0.2306, 0.2049, 0.1965, 0.1979, 0.1737, 0.2189,
0.1984, 0.1790, 0.1813, 0.2076, 0.1848, 0.2157, 0.2131,
0.1944, 0.2044, 0.2646, 0.2174, 0.1902, 0.1935, 0.1748,
0.2613, 0.1727, 0.2092, 0.2226, 0.1858, 0.1944, 0.1705,
0.2480, 0.2006, 0.1727, 0.1548, 0.1953, 0.2012, 0.2029,
0.1774, 0.1911, 0.1929, 0.1892, 0.2119, 0.2406, 0.2135,
0.2534, 0.2343, 0.1954, 0.2331, 0.1886, 0.2186, 0.1790,
0.1965, 0.2009, 0.1768, 0.1803, 0.1938, 0.2262, 0.2026,
0.2061, 0.1688, 0.2048, 0.2277, 0.2105, 0.2118, 0.2376,
0.2225, 0.1665, 0.1702, 0.1705, 0.1555, 0.1686, 0.1897,
0.1924, 0.2212, 0.1598, 0.2279, 0.1994, 0.2152, 0.1557,
0.2102, 0.1926, 0.1695, 0.1386, 0.1956, 0.2050, 0.1709,
0.2047], device='cuda:0')), ('layer4.0.bn2.bias', tensor([-0.0931, -0.1499, -0.0615, -0.1222, -0.0836, -0.0495, -0.1133,
-0.1419, -0.1011, -0.0741, -0.1016, -0.0649, -0.0689, -0.1524,
-0.0230, -0.0677, 0.1640, -0.0709, -0.0411, -0.0881, -0.0307,
-0.1012, -0.0097, -0.0381, -0.0506, -0.1397, -0.0665, -0.1135,
-0.0467, -0.0272, -0.0335, -0.0002, -0.0847, 0.0022, -0.0126,
-0.1373, -0.0167, -0.0758, -0.1466, -0.0610, -0.0655, -0.0105,
-0.0003, -0.0623, -0.1132, -0.0330, -0.1114, -0.0996, -0.1232,
-0.0839, -0.1164, -0.0328, 0.0256, -0.0700, -0.0901, -0.0622,
-0.0676, -0.0880, -0.1016, -0.1158, -0.1477, -0.0989, -0.0560,
-0.0595, -0.1323, -0.0446, -0.0795, -0.0431, -0.1293, -0.1242,
0.0316, -0.1096, -0.1441, -0.0889, -0.0844, -0.0073, -0.0193,
-0.1073, -0.0413, -0.0371, -0.0669, -0.0425, -0.1173, -0.0579,
-0.0932, -0.0409, -0.1212, -0.0618, -0.0060, -0.0903, -0.0859,
-0.0945, -0.0972, -0.1531, 0.1669, -0.0773, -0.0662, -0.1135,
-0.0828, -0.0683, -0.1241, -0.0629, -0.0177, -0.1360, -0.0871,
-0.0848, -0.0598, -0.0979, -0.0614, -0.0804, -0.0421, -0.0444,
-0.1010, -0.0603, -0.0662, -0.0096, -0.0456, -0.1553, 0.0173,
-0.0454, -0.0200, -0.0889, -0.1369, -0.0847, -0.0988, 0.0312,
-0.1040, -0.0377, -0.0961, -0.0451, -0.1223, -0.1104, -0.0443,
-0.0199, -0.0169, -0.0910, -0.0680, -0.1048, -0.1082, -0.1049,
-0.0254, -0.0548, -0.0666, -0.0883, -0.0040, 0.0104, -0.0215,
-0.0268, -0.1044, -0.0775, -0.0302, -0.0593, -0.0869, -0.0996,
-0.0668, -0.1063, -0.1078, -0.0033, -0.0541, -0.1626, -0.0929,
-0.0521, -0.0496, -0.0942, -0.0681, -0.1485, -0.0629, -0.0156,
-0.0636, -0.0697, -0.0857, -0.0841, -0.1379, -0.0259, -0.1447,
-0.1549, -0.0145, -0.0345, -0.0912, -0.0591, -0.0852, -0.0889,
-0.0328, -0.0020, -0.0310, -0.0713, -0.0480, -0.0808, -0.0270,
-0.0059, -0.0300, -0.0930, -0.0236, -0.0335, -0.0165, -0.1012,
0.0156, -0.0961, -0.0640, -0.0513, -0.0749, 0.0066, -0.0112,
-0.0909, -0.0881, -0.0632, -0.1019, -0.1331, -0.0721, -0.0718,
-0.0700, -0.0911, 0.0435, -0.0331, -0.0542, -0.1046, -0.1857,
-0.0869, -0.0866, -0.0572, -0.1017, -0.0519, -0.0490, -0.0123,
-0.0898, -0.0633, -0.0274, -0.0557, -0.0876, -0.0573, -0.1648,
-0.0880, -0.0567, -0.1182, -0.0988, -0.0170, -0.0717, -0.1363,
-0.0966, -0.1085, -0.1143, -0.1261, -0.1013, -0.0838, -0.0361,
-0.0981, -0.0437, -0.0497, -0.0296, -0.0106, -0.0383, -0.0593,
0.0468, -0.0815, -0.0704, -0.0776, -0.1056, -0.0184, -0.0884,
-0.0606, -0.1053, -0.0469, -0.0805, -0.0633, -0.0875, -0.0447,
-0.0193, -0.1020, -0.0569, -0.0948, -0.0266, -0.0203, -0.0661,
-0.1418, -0.0829, -0.1059, -0.0130, -0.0895, -0.1060, -0.0830,
-0.0574, -0.0949, -0.0560, -0.0629, -0.1278, -0.0720, -0.0132,
-0.0683, -0.0460, -0.0389, -0.0813, -0.1214, -0.0453, -0.0905,
-0.0777, -0.1044, -0.0881, -0.0204, -0.0407, -0.0269, -0.0983,
-0.0521, -0.0753, 0.0047, -0.1069, -0.0964, -0.1182, -0.0969,
-0.0482, -0.0449, -0.0709, -0.0776, -0.0786, -0.0457, -0.0178,
-0.0765, -0.1033, -0.0395, -0.0455, -0.0593, -0.0833, -0.0201,
-0.0531, -0.1040, -0.0856, -0.0728, -0.0849, -0.0804, -0.1025,
-0.0630, -0.0742, -0.1253, -0.0017, -0.1053, -0.0326, -0.0972,
-0.0273, -0.0922, -0.0358, -0.0916, -0.1598, -0.0869, -0.0879,
-0.0856, -0.1071, -0.0581, -0.0288, -0.1339, 0.1171, -0.0397,
-0.0744, -0.0870, -0.1716, -0.0737, -0.0213, -0.1089, -0.1708,
-0.0751, -0.0875, 0.0033, -0.0345, -0.1021, -0.0637, -0.0361,
-0.0046, -0.1398, -0.0695, -0.0417, -0.0579, -0.1148, -0.0844,
-0.0890, -0.0367, -0.0158, -0.0915, -0.1197, -0.0158, -0.0752,
-0.1308, -0.0037, -0.0932, -0.0748, -0.0593, -0.1148, -0.0860,
0.0463, -0.0909, -0.1168, -0.1050, -0.0088, -0.1141, -0.1123,
-0.0910, -0.1013, 0.0267, -0.0525, -0.0764, -0.0705, -0.0307,
0.0038, -0.1190, -0.0479, -0.1089, 0.0002, -0.0457, -0.1167,
-0.0555, -0.0192, -0.0807, -0.1642, -0.1093, -0.0172, -0.0761,
-0.0598, -0.1069, -0.1378, -0.0897, -0.0776, -0.0398, -0.0754,
-0.0419, -0.1939, -0.1184, -0.1242, -0.0933, -0.0095, -0.0697,
-0.0894, -0.0912, -0.1053, -0.0925, -0.0741, -0.0641, -0.1039,
-0.0377, -0.0369, -0.0825, -0.1318, -0.0445, -0.1164, -0.0591,
-0.0973, -0.0793, -0.1281, -0.0795, -0.0714, -0.0434, -0.0495,
-0.0872, -0.0331, -0.0576, -0.0698, -0.1018, -0.0124, -0.0807,
-0.1153, -0.0693, -0.0408, -0.0315, -0.0099, -0.0606, -0.1023,
-0.0407, -0.0541, -0.0543, -0.0392, -0.0345, -0.0390, -0.1034,
-0.0938, -0.1134, -0.0950, -0.0866, -0.1148, -0.0780, -0.0344,
-0.1678, -0.0478, -0.0367, -0.1113, -0.0404, -0.1816, -0.0517,
-0.0684, -0.0399, -0.1345, -0.0716, -0.1043, -0.1140, -0.1039,
-0.0735, -0.0397, 0.0072, -0.0627, -0.0277, 0.0104, -0.0551,
-0.1056, -0.0746, -0.0619, -0.0835, -0.0873, -0.0605, -0.0349,
-0.0848, -0.0396, -0.0548, 0.0202, -0.0453, -0.0337, -0.0246,
-0.1700], device='cuda:0')), ('layer4.0.bn2.running_mean', tensor([-0.1257, -0.0933, -0.1496, -0.0753, -0.0810, 0.0439, -0.0648,
-0.1681, -0.0817, -0.0859, -0.0416, -0.0548, -0.1831, -0.0839,
-0.0818, -0.0776, -0.0915, -0.1104, -0.0456, -0.0364, -0.0323,
-0.1026, -0.0951, -0.0924, -0.0327, -0.0876, -0.1300, -0.0507,
-0.1087, -0.0721, -0.0020, -0.0638, -0.0509, -0.0054, -0.0598,
-0.0653, -0.1101, -0.1104, -0.0453, -0.0496, -0.0260, -0.1481,
-0.0576, -0.1055, -0.0562, -0.0819, -0.0767, -0.1357, -0.0509,
-0.0161, -0.0487, -0.0907, 0.0137, -0.0499, -0.0907, -0.0593,
-0.0044, -0.0802, -0.0677, -0.0599, -0.1263, -0.0431, -0.0317,
-0.0055, -0.1518, -0.0031, -0.0777, -0.0181, -0.0527, -0.1314,
0.0046, -0.1661, -0.0284, -0.1753, -0.0826, -0.0054, -0.0372,
-0.1114, -0.0963, -0.0225, -0.0801, -0.0975, -0.0739, -0.1142,
-0.0972, -0.1224, -0.0893, -0.0725, -0.0464, -0.1295, -0.0556,
-0.0576, -0.0845, -0.0292, 0.2877, -0.1139, -0.0710, -0.1203,
-0.0800, -0.0428, -0.0925, 0.0403, -0.0749, -0.0454, -0.1056,
-0.0784, -0.0668, -0.1087, -0.0617, -0.1364, -0.0368, -0.0499,
-0.0525, -0.0157, -0.0693, -0.0511, -0.1523, -0.1062, -0.1319,
-0.0880, -0.0431, -0.0724, -0.0608, -0.0639, -0.0495, -0.1239,
-0.0477, 0.0175, -0.0768, -0.0657, -0.0573, -0.1310, -0.0668,
-0.0398, -0.0293, -0.0578, 0.0792, -0.1000, -0.0673, -0.0939,
-0.0196, -0.0410, -0.0876, -0.0657, -0.0417, -0.0472, -0.0067,
0.0148, -0.1140, -0.0332, 0.0035, -0.0640, -0.1344, -0.0680,
-0.0700, -0.0118, -0.1134, -0.0176, -0.0632, -0.1123, -0.0985,
-0.0993, -0.0657, -0.1091, -0.0894, -0.0727, -0.0362, 0.0063,
-0.0451, -0.0488, -0.0745, -0.0347, -0.1163, -0.0838, -0.1543,
-0.1278, -0.1099, -0.1087, -0.0803, -0.1496, -0.1256, -0.1299,
-0.0619, -0.0920, -0.1153, -0.0446, -0.0783, -0.0933, -0.0609,
-0.0293, -0.1158, -0.0464, -0.0332, -0.0390, -0.0314, -0.0632,
-0.0782, -0.0673, -0.0394, -0.0607, -0.0443, -0.0876, -0.1098,
0.0089, -0.0764, -0.0492, -0.0875, -0.0891, -0.0819, -0.1019,
-0.1154, -0.0805, -0.0733, -0.0617, -0.1182, -0.0564, -0.1103,
-0.0153, -0.0951, -0.0387, -0.0504, -0.1199, -0.0453, -0.0412,
-0.1318, -0.1012, -0.0710, -0.0204, -0.0949, -0.0737, -0.1559,
-0.1152, -0.0797, -0.0779, -0.0186, -0.0589, -0.0811, -0.1381,
-0.1209, -0.1271, -0.0613, -0.0368, -0.0817, -0.0764, -0.0973,
-0.0696, -0.0885, -0.0954, -0.0526, -0.0731, -0.0428, -0.0651,
-0.0551, -0.1161, -0.0432, -0.0622, -0.0489, -0.0934, -0.0819,
-0.0299, -0.0829, -0.0988, -0.0549, -0.1351, -0.1212, -0.0487,
0.0081, -0.1119, -0.1216, -0.1140, -0.0937, -0.0370, -0.0336,
-0.1040, -0.1243, -0.0348, -0.1525, -0.0656, -0.1090, -0.0885,
-0.0496, -0.0972, -0.0655, -0.0721, -0.1047, -0.1336, -0.0903,
-0.0780, -0.0230, -0.0428, -0.0420, -0.0843, -0.1021, -0.0865,
-0.1177, -0.0413, -0.0340, -0.0448, -0.0548, -0.0850, -0.0946,
-0.0297, -0.0749, -0.0527, -0.1130, -0.0717, -0.1128, -0.0491,
-0.0559, -0.0721, -0.0203, -0.1156, -0.1006, 0.0041, -0.0474,
-0.0969, -0.1080, -0.0920, -0.0311, -0.0972, -0.0811, 0.0320,
-0.1531, -0.0496, -0.1172, -0.0608, -0.0449, -0.1234, -0.0905,
-0.0643, -0.0506, -0.1008, -0.0864, -0.1009, -0.0228, -0.0565,
-0.1072, -0.0732, -0.0587, -0.0718, -0.0538, -0.0671, -0.0448,
-0.1225, -0.1111, -0.0281, -0.0784, -0.1247, 0.0753, -0.0961,
-0.0408, -0.0818, -0.1326, -0.1196, -0.0321, -0.1992, -0.0950,
-0.1232, -0.0974, -0.0646, -0.0524, -0.0481, -0.1048, -0.0349,
-0.1079, -0.0894, -0.0054, -0.0768, -0.0311, -0.1383, -0.0576,
-0.0686, -0.0671, -0.0998, -0.0605, -0.1084, -0.1258, -0.0430,
-0.0720, 0.0182, -0.0935, 0.0151, -0.0992, -0.0867, -0.1160,
0.0187, -0.0535, -0.0992, -0.0650, -0.1189, -0.0649, -0.1054,
-0.0998, -0.1159, -0.0537, -0.0088, -0.0006, -0.0959, -0.0196,
-0.0961, -0.0513, -0.0994, -0.0665, -0.0055, -0.0595, -0.1042,
-0.1174, -0.0239, -0.1807, -0.0919, -0.1308, -0.1298, -0.0129,
-0.1006, -0.1181, -0.0549, -0.1062, -0.1371, -0.0276, -0.0187,
-0.0480, -0.1586, -0.0761, -0.0785, -0.1141, -0.0245, -0.0947,
-0.1157, -0.0777, -0.0558, -0.0859, -0.0547, 0.0098, -0.0646,
-0.1291, -0.1152, -0.0338, -0.1176, -0.0787, -0.0941, -0.0842,
-0.0673, -0.0717, -0.1535, -0.0649, -0.1360, -0.0656, -0.0919,
-0.0658, -0.0853, -0.1227, -0.0692, -0.0848, -0.0261, -0.0757,
-0.0311, -0.0727, -0.0186, -0.0405, -0.0361, -0.0793, -0.0497,
0.0390, -0.0970, -0.1521, -0.1177, -0.0446, -0.0758, -0.0787,
-0.0905, -0.0721, -0.0654, -0.0891, -0.0554, -0.0831, -0.0418,
-0.1441, -0.0007, -0.0499, -0.1096, -0.0136, -0.1073, -0.0353,
-0.0244, -0.0728, -0.1080, -0.0677, -0.0418, -0.1010, -0.1386,
-0.1021, -0.1042, -0.0247, -0.1179, -0.0046, -0.0882, -0.1367,
-0.0905, -0.0346, -0.0760, 0.0055, -0.1535, -0.0581, -0.0065,
-0.1126, -0.0061, 0.0226, -0.0810, -0.0854, -0.0299, -0.1797,
-0.0838], device='cuda:0')), ('layer4.0.bn2.running_var', tensor([ 0.0173, 0.0132, 0.0229, 0.0217, 0.0158, 0.0397, 0.0175,
0.0187, 0.0206, 0.0239, 0.0139, 0.0127, 0.0285, 0.0123,
0.0205, 0.0225, 0.0321, 0.0160, 0.0141, 0.0097, 0.0160,
0.0223, 0.0188, 0.0137, 0.0134, 0.0129, 0.0273, 0.0139,
0.0180, 0.0204, 0.0242, 0.0140, 0.0127, 0.0130, 0.0254,
0.0197, 0.0143, 0.0228, 0.0143, 0.0122, 0.0122, 0.0168,
0.0137, 0.0164, 0.0141, 0.0161, 0.0134, 0.0137, 0.0144,
0.0122, 0.0117, 0.0171, 0.0117, 0.0161, 0.0154, 0.0175,
0.0185, 0.0148, 0.0160, 0.0146, 0.0094, 0.0110, 0.0150,
0.0186, 0.0241, 0.0131, 0.0170, 0.0123, 0.0184, 0.0156,
0.0127, 0.0583, 0.0151, 0.0241, 0.0145, 0.0120, 0.0170,
0.0126, 0.0138, 0.0153, 0.0167, 0.0224, 0.0125, 0.0181,
0.0196, 0.0202, 0.0256, 0.0132, 0.0182, 0.0224, 0.0202,
0.0215, 0.0185, 0.0125, 0.2226, 0.0170, 0.0122, 0.0210,
0.0116, 0.0143, 0.0147, 0.0154, 0.0207, 0.0171, 0.0250,
0.0161, 0.0142, 0.0151, 0.0154, 0.0170, 0.0154, 0.0129,
0.0124, 0.0129, 0.0136, 0.0140, 0.0173, 0.0152, 0.0154,
0.0162, 0.0177, 0.0129, 0.0127, 0.0145, 0.0167, 0.0204,
0.0223, 0.0111, 0.0150, 0.0112, 0.0121, 0.0195, 0.0117,
0.0155, 0.0164, 0.0121, 0.0141, 0.0160, 0.0168, 0.0158,
0.0130, 0.0171, 0.0135, 0.0117, 0.0118, 0.0174, 0.0097,
0.0126, 0.0171, 0.0173, 0.0124, 0.0157, 0.0167, 0.0163,
0.0162, 0.0122, 0.0126, 0.0145, 0.0161, 0.0138, 0.0096,
0.0187, 0.0174, 0.0126, 0.0117, 0.0121, 0.0167, 0.0125,
0.0147, 0.0144, 0.0129, 0.0118, 0.0120, 0.0184, 0.0215,
0.0154, 0.0191, 0.0131, 0.0178, 0.0154, 0.0132, 0.0164,
0.0180, 0.0172, 0.0153, 0.0128, 0.0144, 0.0161, 0.0150,
0.0194, 0.0232, 0.0145, 0.0144, 0.0137, 0.0145, 0.0145,
0.0134, 0.0147, 0.0113, 0.0171, 0.0120, 0.0179, 0.0133,
0.0127, 0.0181, 0.0170, 0.0119, 0.0157, 0.0217, 0.0152,
0.0256, 0.0195, 0.0194, 0.0128, 0.0267, 0.0124, 0.0135,
0.0148, 0.0151, 0.0139, 0.0144, 0.0207, 0.0300, 0.0144,
0.0135, 0.0174, 0.0140, 0.0141, 0.0140, 0.0148, 0.0230,
0.0143, 0.0114, 0.0183, 0.0160, 0.0148, 0.0141, 0.0148,
0.0139, 0.0147, 0.0137, 0.0157, 0.0154, 0.0135, 0.0167,
0.0158, 0.0126, 0.0250, 0.0101, 0.0159, 0.0144, 0.0215,
0.0314, 0.0190, 0.0138, 0.0150, 0.0146, 0.0145, 0.0169,
0.0208, 0.0141, 0.0191, 0.0152, 0.0125, 0.0211, 0.0133,
0.0127, 0.0137, 0.0226, 0.0139, 0.0140, 0.0125, 0.0143,
0.0159, 0.0184, 0.0112, 0.0218, 0.0152, 0.0126, 0.0173,
0.0201, 0.0237, 0.0205, 0.0240, 0.0169, 0.0122, 0.0127,
0.0129, 0.0145, 0.0128, 0.0183, 0.0165, 0.0221, 0.0209,
0.0139, 0.0117, 0.0108, 0.0142, 0.0170, 0.0154, 0.0128,
0.0168, 0.0131, 0.0166, 0.0154, 0.0127, 0.0176, 0.0123,
0.0192, 0.0236, 0.0199, 0.0121, 0.0243, 0.0208, 0.0130,
0.0118, 0.0128, 0.0125, 0.0197, 0.0126, 0.0155, 0.0126,
0.0219, 0.0120, 0.0137, 0.0118, 0.0220, 0.0190, 0.0151,
0.0170, 0.0089, 0.0192, 0.0228, 0.0166, 0.0185, 0.0158,
0.0193, 0.0196, 0.0219, 0.0146, 0.0137, 0.0177, 0.0117,
0.0160, 0.0156, 0.0148, 0.0122, 0.0152, 0.0306, 0.0221,
0.0177, 0.0170, 0.0171, 0.0157, 0.0189, 0.0251, 0.0176,
0.0165, 0.0139, 0.0179, 0.0170, 0.0228, 0.0129, 0.0157,
0.0178, 0.0148, 0.0170, 0.0121, 0.0144, 0.0200, 0.0148,
0.0163, 0.0140, 0.0164, 0.0249, 0.0173, 0.0197, 0.0140,
0.0097, 0.0131, 0.0251, 0.0121, 0.0267, 0.0218, 0.0193,
0.0163, 0.0155, 0.0160, 0.0173, 0.0303, 0.0285, 0.0139,
0.0135, 0.0269, 0.0180, 0.0140, 0.0162, 0.0254, 0.0132,
0.0208, 0.0150, 0.0130, 0.0189, 0.0122, 0.0189, 0.0126,
0.0224, 0.0135, 0.0207, 0.0112, 0.0315, 0.0247, 0.0144,
0.0323, 0.0168, 0.0150, 0.0150, 0.0199, 0.0132, 0.0156,
0.0150, 0.0130, 0.0141, 0.0153, 0.0157, 0.0172, 0.0170,
0.0178, 0.0211, 0.0178, 0.0208, 0.0162, 0.0130, 0.0240,
0.0157, 0.0315, 0.0135, 0.0190, 0.0145, 0.0149, 0.0219,
0.0115, 0.0116, 0.0274, 0.0189, 0.0261, 0.0142, 0.0161,
0.0160, 0.0187, 0.0194, 0.0163, 0.0105, 0.0188, 0.0181,
0.0198, 0.0238, 0.0159, 0.0132, 0.0152, 0.0162, 0.0159,
0.0139, 0.0165, 0.0271, 0.0210, 0.0203, 0.0184, 0.0172,
0.0208, 0.0169, 0.0145, 0.0155, 0.0119, 0.0153, 0.0167,
0.0192, 0.0125, 0.0222, 0.0102, 0.0164, 0.0147, 0.0134,
0.0166, 0.0133, 0.0220, 0.0163, 0.0131, 0.0126, 0.0200,
0.0130, 0.0199, 0.0158, 0.0118, 0.0109, 0.0143, 0.0157,
0.0148, 0.0143, 0.0135, 0.0126, 0.0227, 0.0193, 0.0166,
0.0157, 0.0163, 0.0123, 0.0132, 0.0159, 0.0129, 0.0280,
0.0118], device='cuda:0')), ('layer4.0.conv3.weight', tensor([[[[ 4.7218e-04]],
[[ 2.8869e-03]],
[[-1.0411e-02]],
...,
[[-1.8233e-03]],
[[-4.5135e-03]],
[[-1.1556e-02]]],
[[[-4.9109e-03]],
[[ 4.0970e-02]],
[[-1.0509e-02]],
...,
[[-5.5564e-03]],
[[-2.4599e-02]],
[[-2.2074e-03]]],
[[[ 1.3138e-02]],
[[-1.7743e-02]],
[[-7.9763e-03]],
...,
[[ 8.6013e-03]],
[[-3.3060e-03]],
[[-1.5336e-02]]],
...,
[[[ 1.3885e-02]],
[[-1.1164e-02]],
[[-1.0536e-02]],
...,
[[-1.3578e-02]],
[[ 4.1712e-03]],
[[ 1.8974e-02]]],
[[[ 4.8031e-03]],
[[ 1.7265e-03]],
[[-1.4541e-02]],
...,
[[ 1.7751e-02]],
[[-8.2708e-03]],
[[ 5.0612e-04]]],
[[[ 1.7116e-02]],
[[-8.0324e-04]],
[[ 1.3431e-02]],
...,
[[ 8.7292e-03]],
[[ 1.4907e-02]],
[[-1.0531e-02]]]], device='cuda:0')), ('layer4.0.bn3.weight', tensor([ 0.3161, 0.4233, 0.3980, ..., 0.3211, 0.3659, 0.3139], device='cuda:0')), ('layer4.0.bn3.bias', tensor([-0.0554, -0.0932, -0.0689, ..., -0.0629, -0.0640, -0.0678], device='cuda:0')), ('layer4.0.bn3.running_mean', tensor([-0.0132, -0.0295, -0.0231, ..., -0.0178, -0.0218, -0.0000], device='cuda:0')), ('layer4.0.bn3.running_var', tensor(1.00000e-02 *
[ 0.1272, 0.1592, 0.1827, ..., 0.1071, 0.1366, 0.1039], device='cuda:0')), ('layer4.0.downsample.0.weight', tensor([[[[-2.8940e-04]],
[[-5.0691e-03]],
[[-2.1303e-04]],
...,
[[-6.4571e-03]],
[[-5.4951e-03]],
[[ 7.8865e-03]]],
[[[ 7.3810e-04]],
[[ 2.3960e-03]],
[[ 2.2357e-03]],
...,
[[-3.5516e-03]],
[[-7.9180e-03]],
[[-5.7558e-05]]],
[[[-6.4811e-03]],
[[-6.2782e-03]],
[[ 1.3192e-03]],
...,
[[-1.6166e-02]],
[[ 1.1604e-04]],
[[ 5.4752e-04]]],
...,
[[[-2.3182e-03]],
[[ 1.1289e-02]],
[[ 6.1512e-03]],
...,
[[ 1.3922e-03]],
[[ 1.1286e-02]],
[[-1.0417e-02]]],
[[[-2.2377e-03]],
[[ 2.6082e-03]],
[[-9.7219e-03]],
...,
[[-7.2215e-03]],
[[-2.9952e-03]],
[[-1.8826e-03]]],
[[[ 3.1162e-03]],
[[ 1.1647e-04]],
[[-5.6136e-03]],
...,
[[ 1.9132e-02]],
[[-1.1857e-02]],
[[ 1.6298e-02]]]], device='cuda:0')), ('layer4.0.downsample.1.weight', tensor([ 0.2437, 0.2723, 0.2603, ..., 0.1800, 0.2692, 0.2507], device='cuda:0')), ('layer4.0.downsample.1.bias', tensor([-0.0554, -0.0932, -0.0689, ..., -0.0629, -0.0640, -0.0678], device='cuda:0')), ('layer4.0.downsample.1.running_mean', tensor(1.00000e-02 *
[-1.3175, 0.3593, -2.7977, ..., 0.2147, -1.9718, -0.2634], device='cuda:0')), ('layer4.0.downsample.1.running_var', tensor(1.00000e-02 *
[ 0.1292, 0.1668, 0.1731, ..., 0.1122, 0.2180, 0.1686], device='cuda:0')), ('layer4.1.conv1.weight', tensor([[[[-2.1300e-02]],
[[-2.0314e-05]],
[[-1.4222e-02]],
...,
[[-1.0094e-02]],
[[-1.8326e-02]],
[[-7.2067e-03]]],
[[[-1.9785e-03]],
[[ 2.4680e-02]],
[[-5.0996e-03]],
...,
[[ 9.8008e-03]],
[[ 1.7583e-02]],
[[-2.1107e-02]]],
[[[-2.3813e-02]],
[[-8.8521e-03]],
[[-1.5111e-02]],
...,
[[ 8.2568e-04]],
[[ 1.5317e-02]],
[[-4.7038e-04]]],
...,
[[[ 2.0123e-02]],
[[-1.4689e-02]],
[[-1.5127e-02]],
...,
[[ 8.8267e-03]],
[[-1.5475e-03]],
[[ 6.3691e-04]]],
[[[-1.0118e-02]],
[[-2.4782e-02]],
[[-3.3125e-03]],
...,
[[-4.5122e-03]],
[[ 7.3443e-03]],
[[-7.9747e-04]]],
[[[-3.0199e-02]],
[[-1.3252e-02]],
[[ 1.4857e-02]],
...,
[[ 6.6698e-03]],
[[-2.3857e-04]],
[[ 5.0959e-03]]]], device='cuda:0')), ('layer4.1.bn1.weight', tensor([ 0.1845, 0.1796, 0.1810, 0.1958, 0.2202, 0.1341, 0.2056,
0.2023, 0.2450, 0.1650, 0.2402, 0.1890, 0.1999, 0.2175,
0.1728, 0.1884, 0.1275, 0.2154, 0.1501, 0.2084, 0.1932,
0.2435, 0.1551, 0.1732, 0.1896, 0.1658, 0.2160, 0.2082,
0.0949, 0.1605, 0.2024, 0.1793, 0.2575, 0.1878, 0.2150,
0.1432, 0.2199, 0.1996, 0.1665, 0.1983, 0.0820, 0.2224,
0.1884, 0.1719, 0.2083, 0.1816, 0.2119, 0.2163, 0.1926,
0.1456, 0.1886, 0.2340, 0.2138, 0.1284, 0.2306, 0.2006,
0.1921, 0.2096, 0.1715, 0.1830, 0.2158, 0.1866, 0.2040,
0.1900, 0.2141, 0.2107, 0.1891, 0.1550, 0.2144, 0.2186,
0.1646, 0.2137, 0.2517, 0.1622, 0.1864, 0.1773, 0.1898,
0.1677, 0.2133, 0.1278, 0.1896, 0.2556, 0.2032, 0.1698,
0.2248, 0.2036, 0.1996, 0.1868, 0.1952, 0.2144, 0.2291,
0.2174, 0.1906, 0.2099, 0.2196, 0.2058, 0.2304, 0.1763,
0.1297, 0.1660, 0.1548, 0.2107, 0.2059, 0.1183, 0.2271,
0.2020, 0.2136, 0.1772, 0.2196, 0.1656, 0.1877, 0.1914,
0.1994, 0.1641, 0.1878, 0.1638, 0.2461, 0.1845, 0.2452,
0.2021, 0.1673, 0.1903, 0.1439, 0.1956, 0.1757, 0.2206,
0.2011, 0.2283, 0.1545, 0.2252, 0.1797, 0.1482, 0.1747,
0.2029, 0.1489, 0.1434, 0.1716, 0.1545, 0.2313, 0.1358,
0.2008, 0.2106, 0.2025, 0.1614, 0.1599, 0.1957, 0.1552,
0.2501, 0.1859, 0.2232, 0.1696, 0.1325, 0.1613, 0.1852,
0.2309, 0.2386, 0.2143, 0.1864, 0.1806, 0.2045, 0.1988,
0.1716, 0.1905, 0.1740, 0.1845, 0.2257, 0.1537, 0.2106,
0.1764, 0.1460, 0.2010, 0.1657, 0.1694, 0.1656, 0.1958,
0.1857, 0.1906, 0.1632, 0.2103, 0.1898, 0.1800, 0.1520,
0.1695, 0.1761, 0.2114, 0.1723, 0.2061, 0.2180, 0.2183,
0.1728, 0.1815, 0.1876, 0.2014, 0.1887, 0.1785, 0.2228,
0.1985, 0.1708, 0.2332, 0.1802, 0.2007, 0.1636, 0.2238,
0.1964, 0.2295, 0.1895, 0.2269, 0.2294, 0.1749, 0.1765,
0.1967, 0.1717, 0.1940, 0.2127, 0.1813, 0.2138, 0.1398,
0.2053, 0.2074, 0.1994, 0.2142, 0.1711, 0.2500, 0.2440,
0.2050, 0.1607, 0.2139, 0.2188, 0.2385, 0.1948, 0.2335,
0.1823, 0.1505, 0.1924, 0.1399, 0.1872, 0.2081, 0.1687,
0.2906, 0.2088, 0.2140, 0.1635, 0.1562, 0.1892, 0.1724,
0.2172, 0.2117, 0.1768, 0.2018, 0.1849, 0.2067, 0.2054,
0.1796, 0.2023, 0.2093, 0.1877, 0.2276, 0.1783, 0.2006,
0.2003, 0.1941, 0.1430, 0.2007, 0.1913, 0.1923, 0.1765,
0.1846, 0.1959, 0.2032, 0.1851, 0.1754, 0.1660, 0.2090,
0.1862, 0.1875, 0.2222, 0.1771, 0.1894, 0.1810, 0.1296,
0.2145, 0.2299, 0.2006, 0.1541, 0.2029, 0.2208, 0.2086,
0.1545, 0.2080, 0.1871, 0.2070, 0.1925, 0.1683, 0.1981,
0.2258, 0.1984, 0.1935, 0.1594, 0.1182, 0.2458, 0.2158,
0.1958, 0.1837, 0.2071, 0.1751, 0.1921, 0.2262, 0.1771,
0.1841, 0.2103, 0.1589, 0.2120, 0.1966, 0.2077, 0.1688,
0.2401, 0.2132, 0.2272, 0.1820, 0.1888, 0.2242, 0.2072,
0.2133, 0.1842, 0.2244, 0.2144, 0.2094, 0.1472, 0.1904,
0.2505, 0.1961, 0.1915, 0.2096, 0.1926, 0.2013, 0.2519,
0.2210, 0.1636, 0.1541, 0.2312, 0.2353, 0.1998, 0.2424,
0.2245, 0.1422, 0.2303, 0.1834, 0.2074, 0.1814, 0.1725,
0.1591, 0.1779, 0.2134, 0.1888, 0.2138, 0.1774, 0.2028,
0.2246, 0.2108, 0.2297, 0.2224, 0.1488, 0.1975, 0.2101,
0.1694, 0.1842, 0.1827, 0.1648, 0.1515, 0.2242, 0.1967,
0.1933, 0.1855, 0.2052, 0.2189, 0.2025, 0.2215, 0.1849,
0.2069, 0.2511, 0.1905, 0.1824, 0.1895, 0.2050, 0.1989,
0.2144, 0.1478, 0.2769, 0.2463, 0.2438, 0.1974, 0.1921,
0.1924, 0.2038, 0.1892, 0.1797, 0.2478, 0.2311, 0.1758,
0.1450, 0.1806, 0.1057, 0.1998, 0.1965, 0.1975, 0.1982,
0.2343, 0.1740, 0.1969, 0.1682, 0.1557, 0.1766, 0.2206,
0.2071, 0.1946, 0.2192, 0.2665, 0.1622, 0.1781, 0.1740,
0.2103, 0.1679, 0.1522, 0.1791, 0.2402, 0.1344, 0.2247,
0.2137, 0.2240, 0.1996, 0.1257, 0.1675, 0.1982, 0.1632,
0.2399, 0.1458, 0.2389, 0.1944, 0.1741, 0.1964, 0.1490,
0.2109, 0.1488, 0.2170, 0.2322, 0.2634, 0.2163, 0.1934,
0.1812, 0.2093, 0.2001, 0.2021, 0.1514, 0.1281, 0.1894,
0.2028, 0.1815, 0.1335, 0.1884, 0.2230, 0.2182, 0.2152,
0.1962, 0.1810, 0.2215, 0.1701, 0.1923, 0.1241, 0.1919,
0.1097, 0.1779, 0.2237, 0.1936, 0.2309, 0.2491, 0.1804,
0.1628, 0.2242, 0.1871, 0.1586, 0.2232, 0.2182, 0.1975,
0.1977, 0.1978, 0.2690, 0.1507, 0.1704, 0.1632, 0.2215,
0.1689, 0.2392, 0.1593, 0.1833, 0.2078, 0.2187, 0.1672,
0.2099, 0.1418, 0.1809, 0.1660, 0.2182, 0.1804, 0.1843,
0.2105, 0.2122, 0.1557, 0.1860, 0.2544, 0.1981, 0.2683,
0.2078], device='cuda:0')), ('layer4.1.bn1.bias', tensor([-0.1246, -0.1195, -0.0672, -0.1554, -0.1566, -0.0381, -0.1282,
-0.1058, -0.1446, -0.0737, -0.1918, -0.1473, -0.1309, -0.1381,
-0.1350, -0.1483, -0.0877, -0.1454, -0.1077, -0.1915, -0.1253,
-0.1461, -0.0863, -0.1260, -0.1565, -0.1490, -0.1227, -0.0940,
0.0831, -0.1192, -0.1569, -0.1315, -0.2022, -0.0559, -0.1557,
-0.0398, -0.1257, -0.0856, -0.1326, -0.0989, 0.0360, -0.1301,
-0.1367, -0.0917, -0.0862, -0.1699, -0.1650, -0.1498, -0.1117,
-0.0812, -0.0591, -0.2285, -0.1425, -0.0713, -0.2244, -0.1515,
-0.1184, -0.1288, -0.1098, -0.1049, -0.0921, -0.0824, -0.1021,
-0.1403, -0.1847, -0.1419, -0.1516, -0.0085, -0.1998, -0.1666,
-0.0725, -0.1754, -0.1960, -0.1102, -0.1150, -0.1010, -0.1530,
-0.0819, -0.1500, 0.1125, -0.1591, -0.1932, -0.1177, -0.0630,
-0.1713, -0.1487, -0.1576, -0.1432, -0.1269, -0.1060, -0.1774,
-0.1668, -0.1747, -0.1862, -0.2177, -0.1290, -0.1232, -0.1672,
-0.1016, -0.0945, -0.0921, -0.1721, -0.1448, -0.0333, -0.1623,
-0.1099, -0.1322, -0.1334, -0.1935, -0.0897, -0.0711, -0.1527,
-0.1127, -0.1042, -0.0345, -0.0775, -0.1439, -0.1354, -0.1530,
-0.1469, -0.1301, -0.1799, -0.0892, -0.1438, -0.1352, -0.1827,
-0.1474, -0.2216, -0.1308, -0.1847, -0.1199, -0.0745, -0.0743,
-0.1929, -0.0983, -0.0662, -0.1237, -0.0703, -0.1434, -0.0192,
-0.1774, -0.1491, -0.0947, -0.0812, -0.0609, -0.1762, 0.0528,
-0.1683, -0.1463, -0.1601, -0.1057, -0.0301, -0.0703, -0.0726,
-0.1276, -0.1667, -0.1521, -0.1158, -0.1120, -0.1745, -0.1065,
-0.0973, -0.0258, -0.1265, -0.1423, -0.1965, -0.0377, -0.1383,
-0.1876, -0.1048, -0.0892, -0.0416, -0.0858, -0.1156, -0.1244,
-0.0616, -0.1592, -0.0204, -0.1591, -0.1068, -0.1224, -0.0767,
-0.0962, -0.0801, -0.1436, -0.1039, -0.1522, -0.1745, -0.1455,
-0.1669, -0.1052, -0.1673, -0.1570, -0.1222, -0.1057, -0.2149,
-0.0838, -0.0824, -0.1501, -0.0565, -0.1369, -0.0638, -0.2023,
-0.1245, -0.1731, -0.1341, -0.1697, -0.1059, -0.0934, -0.0967,
-0.1074, -0.1177, -0.1597, -0.1463, -0.1327, -0.1667, -0.0673,
-0.1950, -0.1291, -0.0587, -0.1557, -0.1157, -0.2025, -0.1375,
-0.1538, -0.0925, -0.0832, -0.1452, -0.2370, -0.1767, -0.1570,
-0.0978, -0.0777, -0.1702, -0.1019, -0.1153, -0.1494, -0.0709,
-0.3220, -0.1513, -0.1740, -0.0655, -0.0427, -0.1431, -0.1392,
-0.1394, -0.1366, -0.0904, -0.1452, -0.0834, -0.0893, -0.1621,
0.1615, -0.1505, -0.1498, -0.1741, -0.1576, -0.1586, -0.1273,
-0.1414, -0.1246, -0.0547, -0.1060, -0.1309, -0.1145, -0.0504,
-0.1064, -0.1471, -0.1888, -0.1585, -0.0917, -0.1517, -0.1230,
-0.1175, -0.1571, -0.1680, -0.0561, -0.1268, -0.0883, -0.0568,
-0.1544, -0.1481, -0.1852, -0.0079, -0.1407, -0.1259, -0.1357,
-0.0701, -0.0485, -0.0687, -0.1333, -0.1389, -0.1488, -0.0978,
-0.1014, -0.1286, -0.1538, -0.1280, -0.0002, -0.1842, -0.1577,
-0.1502, -0.1466, -0.1648, -0.1055, -0.1894, -0.1748, -0.0887,
-0.1085, -0.1528, -0.0706, -0.1180, -0.0656, -0.1491, -0.1208,
-0.1811, -0.1590, -0.1468, -0.1211, -0.1865, -0.1595, -0.1954,
-0.1147, -0.1705, -0.1253, -0.1146, -0.1188, -0.0734, -0.0328,
-0.1500, -0.1561, -0.0896, -0.1735, -0.1923, -0.1544, -0.1202,
-0.1713, -0.0627, -0.1048, -0.2271, -0.1938, -0.1771, -0.1192,
-0.1255, -0.1117, -0.1793, -0.1841, -0.0616, -0.1089, -0.1188,
-0.1437, -0.1289, -0.1832, -0.1454, -0.1635, -0.0711, -0.1026,
-0.1939, -0.0802, -0.1867, -0.1416, -0.0948, -0.1568, -0.1421,
-0.1531, -0.1902, -0.0629, -0.1569, -0.1111, -0.1475, -0.1735,
-0.1369, -0.1359, -0.1511, -0.1509, -0.1201, -0.1236, -0.1339,
-0.0994, -0.1258, -0.0608, -0.1945, -0.1366, -0.1125, -0.1132,
-0.1567, -0.0334, -0.0038, -0.1258, -0.1929, -0.1286, -0.1471,
-0.0721, -0.2075, -0.0704, -0.0737, -0.1399, -0.1354, -0.0960,
-0.0896, -0.1558, -0.0064, -0.1251, -0.0944, -0.1180, -0.1214,
-0.2222, -0.1378, -0.1055, -0.1364, -0.0601, -0.0855, -0.1200,
-0.1212, -0.1767, -0.1484, -0.2268, -0.0802, -0.1748, -0.1194,
-0.1468, -0.0529, -0.0975, -0.1052, -0.1290, -0.1004, -0.1493,
-0.1850, -0.1658, -0.1023, 0.0020, -0.0522, -0.1194, -0.0426,
-0.2160, -0.1091, -0.1873, -0.1522, -0.1014, -0.2206, -0.0146,
-0.1456, -0.1070, -0.1661, -0.1689, -0.1929, -0.1606, -0.1319,
-0.0997, -0.1330, -0.1664, -0.1353, -0.1172, -0.0636, -0.0698,
-0.1356, -0.1474, 0.0068, -0.1643, -0.1430, -0.1643, -0.1310,
-0.1602, -0.1294, -0.1345, -0.1158, -0.1087, -0.0659, -0.1059,
0.1131, -0.1243, -0.1978, -0.0994, -0.1712, -0.1912, -0.1372,
-0.0819, -0.1712, -0.1241, -0.0703, -0.1738, -0.1593, -0.1324,
-0.1418, -0.1975, -0.1508, -0.1097, -0.0913, -0.0350, -0.1510,
-0.1243, -0.1872, -0.0298, -0.1057, -0.1740, -0.1329, -0.1180,
-0.1362, -0.0900, -0.1879, -0.0221, -0.0615, -0.1138, -0.1150,
-0.1244, -0.1450, -0.1208, -0.1237, -0.2752, -0.1437, -0.1871,
-0.1753], device='cuda:0')), ('layer4.1.bn1.running_mean', tensor([-0.2708, -0.1303, -0.5119, -0.2707, -0.1532, -0.1380, -0.6791,
-0.3497, -0.2257, -0.3257, -0.2862, -0.4220, -0.2254, -0.5082,
-0.4364, -0.4032, -0.4614, -0.2719, -0.1467, -0.2444, -0.4286,
-0.5561, -0.2545, -0.4438, -0.5653, -0.3441, -0.0981, -0.4580,
0.0577, -0.1707, -0.3773, -0.1698, -0.5481, -0.4181, -0.0462,
-0.3732, -0.5251, -0.3421, -0.1456, -0.1678, -0.6710, -0.0386,
-0.4590, -0.3435, -0.1328, -0.3852, -0.2289, -0.0490, -0.2470,
-0.3631, -0.1960, -0.3935, -0.1840, -0.1260, -0.2911, -0.4235,
-0.2234, -0.2677, -0.3429, -0.2994, -0.3729, -0.4652, -0.0433,
-0.2028, -0.1404, -0.3016, -0.2854, -0.4372, -0.2663, -0.3925,
-0.1443, -0.3295, -0.3432, -0.2429, -0.1031, -0.1544, -0.1262,
-0.1150, -0.4778, -0.7368, -0.4280, -0.4495, -0.2871, -0.0474,
-0.2052, -0.3349, -0.3821, -0.3485, -0.3074, -0.2242, -0.3506,
-0.1046, -0.2483, -0.3539, -0.3061, -0.2937, -0.4135, -0.2286,
-0.3776, -0.3067, -0.2122, -0.1977, -0.5309, 0.0363, -0.2279,
-0.3101, -0.3100, -0.4388, -0.9304, -0.2811, -0.3706, -0.4536,
-0.1946, -0.3743, -0.1476, -0.2911, -0.3240, -0.1071, -0.5011,
-0.2516, -0.3210, -0.6414, -0.0072, -0.3049, -0.4794, -0.1770,
-0.1524, -0.3806, -0.2646, -0.2571, -0.0604, -0.3388, -0.4292,
-0.3056, 0.1066, -0.2632, -0.4267, -0.3376, -0.4148, -0.2981,
-0.3315, -0.2590, -0.3661, -0.4898, -0.1548, -0.3152, -0.1069,
-0.3533, -0.2237, -0.2209, -0.1282, -0.4776, 0.0280, -0.3389,
-0.3026, -0.1404, -0.4634, -0.1894, -0.3218, -0.1711, -0.1119,
-0.2740, -0.0912, -0.3157, -0.2013, -0.0363, -0.0343, -0.3488,
-0.1924, -0.4278, -0.3171, -0.2053, -0.0907, -0.1918, -0.2765,
-0.3637, -0.2648, -0.5354, -0.2226, -0.1597, -0.2818, -0.2492,
-0.4287, -0.4715, -0.6010, -0.1427, -0.2994, -0.2195, -0.1932,
-0.3544, -0.0402, -0.1884, -0.1753, -0.1119, -0.6109, -0.2265,
-0.4384, -0.4531, -0.2475, -0.6980, -0.3975, 0.0202, -0.3519,
-0.2320, -0.5750, -0.0119, -0.3618, -0.2846, -0.2022, -0.3153,
-0.3766, -0.4625, -0.3034, -0.2305, -0.3115, -0.3019, -0.4783,
-0.2207, -0.1896, -0.3540, -0.2035, -0.5033, -0.3795, -0.6154,
-0.1463, -0.5935, -0.2985, -0.3259, -0.2549, -0.5396, -0.2245,
-0.1355, -0.5182, -0.1460, -0.0546, -0.2032, -0.3829, -0.2261,
-0.4985, -0.3263, -0.1562, -0.4273, -0.2837, -0.3246, -0.3299,
-0.1310, -0.5024, -0.5825, -0.1418, -0.1541, -0.3994, -0.5795,
-0.8513, -0.0505, -0.2693, -0.4677, -0.3132, -0.3560, -0.3421,
-0.2941, -0.3913, 0.0212, -0.4101, -0.2056, -0.3474, -0.4485,
-0.1272, -0.3253, -0.0784, -0.1353, -0.3112, -0.5504, -0.4445,
-0.0596, -0.1572, -0.1819, -0.3935, -0.1131, -0.1856, -0.2800,
-0.4222, -0.3633, -0.2082, -0.2456, -0.1547, -0.1200, -0.2956,
-0.4687, -0.3256, -0.2947, -0.3102, -0.1998, -0.3211, -0.3573,
-0.4035, -0.2771, -0.2501, -0.4308, -0.3355, -0.6267, -0.3640,
-0.3242, -0.3020, -0.4676, -0.1012, -0.2682, -0.2528, -0.1328,
-0.1993, -0.3230, -0.1817, -0.1757, -0.4883, -0.2594, -0.6535,
-0.3116, -0.3610, -0.0907, -0.1469, -0.2187, -0.4639, -0.3700,
-0.2462, -0.3965, -0.3513, -0.3658, -0.3117, -0.0335, -0.4977,
-0.2943, -0.1711, -0.5140, -0.1910, -0.2170, -0.3064, -0.3751,
-0.2782, -0.4345, -0.5462, -0.2852, -0.2446, -0.5553, -0.3794,
-0.1378, -0.4820, -0.1685, -0.3654, -0.2646, -0.3714, -0.1766,
-0.4904, -0.1231, -0.4988, -0.7103, -0.2442, -0.2703, -0.1778,
-0.3654, -0.2025, -0.2980, -0.4395, -0.2835, -0.1940, -0.4494,
-0.2540, -0.5029, -0.2766, -0.4507, -0.4046, -0.2826, -0.3096,
-0.4667, -0.4431, -0.2102, -0.2134, -0.4381, -0.2170, -0.4284,
-0.4092, -0.4134, -0.3294, -0.5260, -0.3005, -0.3410, -0.2317,
-0.1788, -0.4739, 1.0654, -0.3776, -0.3998, -0.2078, -0.1682,
-0.1718, -0.3369, -0.2008, -0.2762, -0.1815, -0.4247, -0.2092,
-0.2727, -0.5458, -0.2403, -0.5369, -0.2811, -0.1961, -0.4054,
-0.3199, -0.4462, -0.4686, -0.3520, -0.3305, -0.3098, -0.0889,
-0.1954, -0.3047, -0.3848, -0.5296, -0.1500, -0.3390, -0.1278,
-0.5696, -0.3666, -0.2582, -0.0458, -0.5011, -0.3610, -0.4130,
-0.0940, -0.1160, -0.4499, -0.4177, -0.2660, -0.3824, -0.1169,
-0.5087, -0.2770, -0.1719, -0.0592, -0.0822, -0.4579, -0.2597,
-0.2545, 0.1427, -0.1937, -0.3072, -0.2575, -0.3781, 0.0029,
-0.2381, -0.1797, -0.2839, -0.2358, -0.2678, -0.3117, -0.2094,
-0.2296, -0.3873, -0.3985, -0.1786, -0.4087, -0.1806, -0.3123,
-0.3537, -0.2050, -0.2291, -0.4034, -0.5287, -0.3994, -0.3602,
-0.0054, -0.3181, -0.1440, -0.1964, 0.0217, -0.3554, -0.1888,
-0.4496, -0.2263, -0.1690, -0.2778, -0.0078, 0.0664, -0.3808,
-0.2273, -0.5756, -0.4714, -0.5528, -0.0503, -0.4975, -0.2664,
-0.1165, -0.2009, -0.3626, -0.3335, -0.4221, -0.3250, -0.5182,
-0.4364, -0.3433, -0.2994, -0.2655, -0.6802, -0.1442, -0.4387,
-0.4439, -0.4421, -0.3574, -0.2273, -0.5937, -0.3299, -0.3496,
-0.1596], device='cuda:0')), ('layer4.1.bn1.running_var', tensor([ 0.1557, 0.1236, 0.1959, 0.1315, 0.1131, 0.1816, 0.2322,
0.1494, 0.1787, 0.1685, 0.1240, 0.1851, 0.1407, 0.1673,
0.1339, 0.1352, 0.1668, 0.1474, 0.1133, 0.1160, 0.1833,
0.2478, 0.1479, 0.1341, 0.1368, 0.1628, 0.1290, 0.1721,
0.3485, 0.1479, 0.1349, 0.1243, 0.1477, 0.1622, 0.1130,
0.1644, 0.1596, 0.1247, 0.1471, 0.1141, 0.3174, 0.1557,
0.1609, 0.1873, 0.1711, 0.1194, 0.1184, 0.0934, 0.1407,
0.1607, 0.1365, 0.1078, 0.1248, 0.1618, 0.1093, 0.1326,
0.1229, 0.1113, 0.1704, 0.1598, 0.1590, 0.1908, 0.2346,
0.1048, 0.0937, 0.1545, 0.1460, 0.1511, 0.0983, 0.1295,
0.1556, 0.1199, 0.1363, 0.1511, 0.1463, 0.1538, 0.1159,
0.1688, 0.1221, 0.3151, 0.1493, 0.1443, 0.1278, 0.1685,
0.1392, 0.1074, 0.1282, 0.1370, 0.1129, 0.1227, 0.1345,
0.1508, 0.1006, 0.1147, 0.1500, 0.1150, 0.1519, 0.0971,
0.1598, 0.1582, 0.1300, 0.1437, 0.1962, 0.1633, 0.1680,
0.1390, 0.1637, 0.1357, 0.2719, 0.1792, 0.1862, 0.1990,
0.1632, 0.1101, 0.1771, 0.1892, 0.1674, 0.1475, 0.1740,
0.1044, 0.1101, 0.1660, 0.1548, 0.1155, 0.1631, 0.1390,
0.1468, 0.1315, 0.1282, 0.1416, 0.1368, 0.1432, 0.1402,
0.1365, 0.1187, 0.1354, 0.1592, 0.1681, 0.1690, 0.2058,
0.1539, 0.1298, 0.1591, 0.2264, 0.1464, 0.1257, 0.3278,
0.1510, 0.1168, 0.1385, 0.1469, 0.2099, 0.1890, 0.1646,
0.1363, 0.1323, 0.1812, 0.1470, 0.1778, 0.1176, 0.1683,
0.1448, 0.1691, 0.1619, 0.1626, 0.1095, 0.1555, 0.1610,
0.1194, 0.1701, 0.1403, 0.1568, 0.1292, 0.1622, 0.1190,
0.1567, 0.1220, 0.2018, 0.1214, 0.1232, 0.1159, 0.1532,
0.1589, 0.2347, 0.1751, 0.1602, 0.1190, 0.1572, 0.1189,
0.1309, 0.1275, 0.1385, 0.1066, 0.1320, 0.1457, 0.1263,
0.2145, 0.2040, 0.1296, 0.2182, 0.1549, 0.1727, 0.1081,
0.1273, 0.1946, 0.1229, 0.1642, 0.1509, 0.1351, 0.1252,
0.1601, 0.1896, 0.1602, 0.1389, 0.1279, 0.1202, 0.1890,
0.0929, 0.1213, 0.1394, 0.1440, 0.2370, 0.1144, 0.1475,
0.1290, 0.2069, 0.1452, 0.1695, 0.1039, 0.1510, 0.1360,
0.1081, 0.1514, 0.1133, 0.1584, 0.1048, 0.1906, 0.1673,
0.1174, 0.1210, 0.1640, 0.1744, 0.1749, 0.1780, 0.1556,
0.1377, 0.1946, 0.1626, 0.1084, 0.1281, 0.1275, 0.1724,
0.4968, 0.1087, 0.1198, 0.1188, 0.1348, 0.1293, 0.1168,
0.1823, 0.1217, 0.1647, 0.1639, 0.1190, 0.2049, 0.1535,
0.1543, 0.1087, 0.1086, 0.1048, 0.1305, 0.1901, 0.2464,
0.1305, 0.0900, 0.1399, 0.1758, 0.1143, 0.1874, 0.1560,
0.1423, 0.2025, 0.1062, 0.2183, 0.1347, 0.1889, 0.1224,
0.1209, 0.1668, 0.1845, 0.1412, 0.1458, 0.1637, 0.1455,
0.1385, 0.1313, 0.1012, 0.1556, 0.1395, 0.1643, 0.1386,
0.1270, 0.1731, 0.1977, 0.1128, 0.1080, 0.1566, 0.1185,
0.1271, 0.1743, 0.1595, 0.1255, 0.2007, 0.1464, 0.2049,
0.1184, 0.1233, 0.1475, 0.1287, 0.1320, 0.1487, 0.1360,
0.1848, 0.1226, 0.1750, 0.1419, 0.1548, 0.1466, 0.1947,
0.1578, 0.1220, 0.1662, 0.1714, 0.1397, 0.1754, 0.2000,
0.2007, 0.1802, 0.1853, 0.1171, 0.1354, 0.1705, 0.1714,
0.1410, 0.1807, 0.1407, 0.1102, 0.1891, 0.1114, 0.1570,
0.1569, 0.1478, 0.1327, 0.2169, 0.1368, 0.1606, 0.1280,
0.1491, 0.1605, 0.1135, 0.1738, 0.1457, 0.1102, 0.1389,
0.1466, 0.1203, 0.1758, 0.1630, 0.1165, 0.1436, 0.1287,
0.1567, 0.1774, 0.1520, 0.1412, 0.1476, 0.1516, 0.2216,
0.1612, 0.2049, 0.1341, 0.1440, 0.1750, 0.1610, 0.1310,
0.1174, 0.2391, 0.3791, 0.1742, 0.1196, 0.1484, 0.1098,
0.1174, 0.1354, 0.1490, 0.1308, 0.1255, 0.1437, 0.1411,
0.1507, 0.1391, 0.1858, 0.1485, 0.1293, 0.1873, 0.2176,
0.1226, 0.1589, 0.2669, 0.1431, 0.1707, 0.1831, 0.1453,
0.1323, 0.0951, 0.1438, 0.1383, 0.1122, 0.0955, 0.1452,
0.1603, 0.1591, 0.1493, 0.1401, 0.1741, 0.1207, 0.1399,
0.1392, 0.1405, 0.1764, 0.2596, 0.1394, 0.1379, 0.1451,
0.1558, 0.1058, 0.1521, 0.1217, 0.1264, 0.1769, 0.1899,
0.1329, 0.1107, 0.1186, 0.1114, 0.1550, 0.1575, 0.1289,
0.1671, 0.1812, 0.1131, 0.1369, 0.1071, 0.2065, 0.1388,
0.1590, 0.1127, 0.2063, 0.1473, 0.1235, 0.1688, 0.1270,
0.1444, 0.1118, 0.1707, 0.1282, 0.2246, 0.1502, 0.1331,
0.2350, 0.1012, 0.1150, 0.2199, 0.1191, 0.1472, 0.1366,
0.1449, 0.1182, 0.1608, 0.1307, 0.1462, 0.1080, 0.1171,
0.1149, 0.1649, 0.1626, 0.1835, 0.1270, 0.1485, 0.1365,
0.1151, 0.0987, 0.1774, 0.2197, 0.1319, 0.1464, 0.1265,
0.1403, 0.1472, 0.1238, 0.1227, 0.2230, 0.1262, 0.1432,
0.1711, 0.1263, 0.1119, 0.1168, 0.1374, 0.1839, 0.1207,
0.1006], device='cuda:0')), ('layer4.1.conv2.weight', tensor([[[[-6.3251e-03, -6.1843e-03, -3.2490e-03],
[-1.2556e-02, -1.6035e-02, -4.6046e-03],
[ 4.7581e-03, -2.8236e-03, -1.6110e-02]],
[[ 5.6100e-03, -6.5169e-03, -5.8682e-03],
[-1.7835e-02, -4.5112e-03, 2.3802e-02],
[-5.3904e-03, 2.9851e-03, 8.8139e-03]],
[[ 3.1901e-02, 6.9846e-04, 1.0832e-02],
[ 2.0183e-02, -1.6786e-02, -8.3026e-03],
[ 1.3473e-02, -1.2425e-02, -8.7970e-03]],
...,
[[ 6.1722e-03, 3.3359e-03, -5.8917e-03],
[-4.3866e-03, 1.2697e-02, 7.3854e-03],
[ 8.1003e-03, 1.7285e-02, 2.0107e-03]],
[[-1.5052e-02, 3.1398e-03, 1.0975e-02],
[-1.0645e-02, -8.8044e-03, 1.9737e-02],
[-3.6782e-03, 4.1974e-03, 1.1501e-02]],
[[-2.7631e-03, 8.9629e-04, 1.8902e-02],
[ 7.7906e-03, 6.6125e-03, 1.2847e-02],
[-5.0630e-03, 1.4801e-02, 2.1284e-02]]],
[[[ 2.0190e-03, -3.8205e-03, -5.4967e-03],
[ 1.0782e-02, 2.1576e-03, 4.7123e-03],
[ 2.3979e-03, 2.3090e-03, -1.6435e-03]],
[[-1.0783e-02, -9.9896e-03, -1.2137e-02],
[-1.8844e-02, -1.3004e-02, -1.4344e-02],
[-7.7846e-03, -1.6745e-03, -1.4481e-02]],
[[ 1.1066e-02, 1.2509e-02, 1.5430e-02],
[ 1.2928e-02, 1.0798e-02, 1.1219e-02],
[ 1.1216e-02, -1.8652e-03, 3.1755e-03]],
...,
[[-8.2659e-03, -6.0146e-03, -2.9159e-03],
[-1.4243e-03, -1.2439e-03, 6.2540e-04],
[-3.8462e-03, -4.9230e-03, -1.5037e-02]],
[[-5.2752e-03, -1.3626e-02, -5.3632e-03],
[-1.7069e-04, -1.9965e-02, -6.2033e-03],
[-1.0487e-02, -1.6036e-02, -1.3327e-02]],
[[-5.5631e-03, -4.0050e-03, -1.4024e-03],
[ 1.8784e-02, 1.9155e-02, 1.4035e-02],
[ 1.0137e-02, 1.2539e-02, 7.1149e-03]]],
[[[-2.3009e-03, -1.0310e-02, -9.4162e-03],
[-8.3223e-03, -1.0427e-02, -2.1453e-02],
[-1.7802e-02, -1.6308e-02, -1.8214e-02]],
[[ 1.1440e-03, 1.0702e-02, 8.0583e-03],
[ 3.1181e-04, -1.0946e-03, 1.0740e-02],
[ 8.1485e-03, 8.2424e-03, 1.3631e-02]],
[[-1.1734e-02, -1.1631e-02, -1.3715e-02],
[-8.7037e-03, -2.7108e-02, -2.0680e-02],
[-1.3425e-02, -2.9020e-02, -1.0489e-02]],
...,
[[ 5.4317e-03, 5.5567e-03, 1.5376e-03],
[ 8.6967e-03, 3.5803e-03, -3.0624e-03],
[ 3.1268e-03, 9.4990e-03, -6.9712e-03]],
[[-1.2484e-02, -1.2009e-02, -9.2159e-04],
[-1.3418e-03, -2.2876e-03, 1.2933e-02],
[-6.3033e-03, -4.4610e-03, -1.1646e-03]],
[[ 6.2315e-03, 6.6485e-03, 3.5573e-03],
[ 5.4496e-03, 1.7982e-02, 4.9796e-03],
[-6.4502e-03, -3.7647e-03, -1.1870e-02]]],
...,
[[[ 1.3085e-02, 1.7154e-02, 2.0488e-02],
[ 1.0662e-03, 1.5197e-02, 1.0913e-02],
[-2.4161e-03, 1.0451e-03, 3.3205e-03]],
[[ 1.7579e-03, -1.5447e-03, 2.9409e-03],
[ 4.5914e-03, 1.3953e-04, 1.1250e-02],
[-4.1265e-03, -3.3971e-03, 2.3011e-03]],
[[ 9.0088e-04, 1.8927e-03, 4.7284e-03],
[-5.9859e-03, 1.0108e-02, 3.7050e-03],
[-6.5329e-03, -2.3316e-03, 6.5436e-07]],
...,
[[-2.4862e-02, -2.5605e-02, -3.2080e-02],
[-2.6093e-02, -2.6446e-02, -3.1854e-02],
[-2.1405e-02, -1.4925e-02, -2.4022e-02]],
[[-1.7519e-02, -6.1521e-03, -1.8938e-02],
[-6.8782e-03, 3.4916e-03, -1.4643e-02],
[-1.6982e-02, -1.5705e-02, -2.3746e-02]],
[[-3.8927e-03, -9.7105e-03, -1.7914e-03],
[ 6.2108e-04, 1.7291e-03, 4.5722e-03],
[ 3.9706e-03, -3.4752e-03, -6.0086e-03]]],
[[[ 3.7137e-03, 2.4114e-03, -9.9020e-04],
[ 4.7176e-03, 4.2579e-03, 1.0346e-03],
[-2.5402e-03, 1.1310e-04, -4.2561e-04]],
[[ 3.6461e-03, 7.0547e-03, 5.5834e-03],
[-6.8321e-03, 1.8982e-03, -5.5329e-03],
[-4.2561e-04, -1.1680e-03, 8.3876e-04]],
[[ 5.9613e-03, -1.3125e-03, 6.0573e-03],
[ 2.8431e-02, 6.9059e-03, 3.2437e-02],
[ 3.1090e-02, 1.5772e-02, 2.3143e-02]],
...,
[[ 2.0601e-03, 1.0378e-02, 4.1638e-03],
[ 6.4715e-03, -2.1595e-03, -5.8891e-03],
[ 7.7895e-04, 6.0295e-03, -1.6358e-03]],
[[ 7.2798e-03, 3.4441e-03, 6.7976e-03],
[ 4.8433e-03, 1.3105e-02, 3.0040e-03],
[ 1.0330e-02, 1.6081e-02, 1.5691e-02]],
[[-5.0896e-03, -7.1577e-04, -2.2885e-03],
[-6.4838e-04, -6.4158e-03, -9.4565e-04],
[-9.7655e-03, 3.8239e-04, 3.3144e-04]]],
[[[-7.4100e-03, -2.3550e-02, -1.7789e-02],
[-2.1356e-02, -1.7157e-02, -2.3011e-02],
[-2.2987e-02, -1.9127e-02, -2.5196e-02]],
[[-8.4506e-03, -1.1625e-02, -8.5567e-03],
[-4.3722e-03, -4.8001e-03, -9.3095e-03],
[ 5.7622e-03, 6.8214e-03, 5.8068e-03]],
[[-2.9115e-02, -9.2254e-03, -1.9183e-02],
[-2.1671e-02, 9.5732e-03, -1.4914e-02],
[-2.5943e-02, -1.5880e-02, -2.5241e-02]],
...,
[[ 1.3371e-02, 1.4983e-02, 1.8938e-02],
[ 2.1412e-02, 1.4455e-02, 2.1238e-02],
[ 8.5844e-03, 1.3523e-02, 1.3173e-02]],
[[-4.9798e-03, 5.4822e-04, 1.5736e-03],
[-1.4768e-03, -1.3585e-02, -4.0981e-03],
[-1.7631e-03, -6.7340e-03, 2.2981e-03]],
[[-1.4815e-03, -6.1545e-03, -6.9575e-03],
[ 2.2389e-03, 4.2356e-03, 2.0663e-03],
[ 5.4872e-03, 1.2546e-02, -3.6466e-03]]]], device='cuda:0')), ('layer4.1.bn2.weight', tensor([ 0.2162, 0.2220, 0.1827, 0.2551, 0.2208, 0.2321, 0.1652,
0.1875, 0.2053, 0.1785, 0.2274, 0.2212, 0.1940, 0.2146,
0.1901, 0.1977, 0.2267, 0.2653, 0.2048, 0.1872, 0.1788,
0.2136, 0.1942, 0.1925, 0.1947, 0.1953, 0.1822, 0.1630,
0.2407, 0.2194, 0.2039, 0.2573, 0.2379, 0.1661, 0.2260,
0.1621, 0.1946, 0.1863, 0.1973, 0.2267, 0.2313, 0.1756,
0.2010, 0.2395, 0.2005, 0.1969, 0.2138, 0.2028, 0.1836,
0.2324, 0.2389, 0.1712, 0.2227, 0.2218, 0.2009, 0.1685,
0.2021, 0.2814, 0.2156, 0.2477, 0.1952, 0.2204, 0.2158,
0.1826, 0.1434, 0.2534, 0.2110, 0.2046, 0.1976, 0.2145,
0.1972, 0.2067, 0.1954, 0.2347, 0.1691, 0.2165, 0.1823,
0.1898, 0.1825, 0.2383, 0.2339, 0.2252, 0.2013, 0.1914,
0.1959, 0.2004, 0.2316, 0.2166, 0.2020, 0.1997, 0.1969,
0.2356, 0.2375, 0.2430, 0.2178, 0.2085, 0.2088, 0.2442,
0.2300, 0.1938, 0.1932, 0.1957, 0.2675, 0.1990, 0.1758,
0.1595, 0.2134, 0.1963, 0.2173, 0.2286, 0.2435, 0.1953,
0.2516, 0.2193, 0.2229, 0.1918, 0.2345, 0.1817, 0.2170,
0.2447, 0.2037, 0.2064, 0.1883, 0.2058, 0.2933, 0.1834,
0.2016, 0.2049, 0.2208, 0.2326, 0.1638, 0.2136, 0.2436,
0.2038, 0.2123, 0.2584, 0.2254, 0.2095, 0.2085, 0.2503,
0.2129, 0.2542, 0.2009, 0.2116, 0.2131, 0.1813, 0.2196,
0.1966, 0.2357, 0.1883, 0.2309, 0.1276, 0.2118, 0.2064,
0.2462, 0.2247, 0.2220, 0.1836, 0.1888, 0.2150, 0.1777,
0.2656, 0.1914, 0.2264, 0.2300, 0.2032, 0.2286, 0.2437,
0.2562, 0.2290, 0.2227, 0.1915, 0.1960, 0.1905, 0.2033,
0.2312, 0.1961, 0.2260, 0.2609, 0.1900, 0.2075, 0.2288,
0.2014, 0.2255, 0.1921, 0.2143, 0.1999, 0.2029, 0.2112,
0.2099, 0.2328, 0.1415, 0.1891, 0.1899, 0.2040, 0.1867,
0.2186, 0.2287, 0.2185, 0.1843, 0.2098, 0.2333, 0.2483,
0.2209, 0.2271, 0.2017, 0.1924, 0.2449, 0.2042, 0.2127,
0.1974, 0.2019, 0.1466, 0.1970, 0.2242, 0.1957, 0.1844,
0.2301, 0.2258, 0.2208, 0.2154, 0.1824, 0.2531, 0.2180,
0.2293, 0.2313, 0.1781, 0.2117, 0.2218, 0.2265, 0.2222,
0.2147, 0.2569, 0.1821, 0.2131, 0.2130, 0.1734, 0.1843,
0.2160, 0.2229, 0.2082, 0.2021, 0.1945, 0.2281, 0.1994,
0.2043, 0.2175, 0.1806, 0.2176, 0.1998, 0.2134, 0.2024,
0.2336, 0.2043, 0.1788, 0.2519, 0.2208, 0.2252, 0.2053,
0.2062, 0.2080, 0.1997, 0.2133, 0.2062, 0.1824, 0.2259,
0.1898, 0.2322, 0.2142, 0.2023, 0.2435, 0.1895, 0.1966,
0.2145, 0.2396, 0.1852, 0.1918, 0.1988, 0.2074, 0.1787,
0.2123, 0.2136, 0.1883, 0.2032, 0.1994, 0.2184, 0.2168,
0.2228, 0.1915, 0.2001, 0.2433, 0.2161, 0.2065, 0.2115,
0.2133, 0.2914, 0.2392, 0.1842, 0.1921, 0.2225, 0.1618,
0.2018, 0.1658, 0.2017, 0.1796, 0.2158, 0.1872, 0.2193,
0.1986, 0.2077, 0.1889, 0.2218, 0.2254, 0.1714, 0.1912,
0.1149, 0.1910, 0.2456, 0.1831, 0.2284, 0.2126, 0.2181,
0.2286, 0.1991, 0.2018, 0.1968, 0.2346, 0.2646, 0.1966,
0.1635, 0.2280, 0.2100, 0.1977, 0.1875, 0.2481, 0.2364,
0.2755, 0.2102, 0.1987, 0.1815, 0.2508, 0.2221, 0.2031,
0.2145, 0.1986, 0.2380, 0.2529, 0.2428, 0.2461, 0.2060,
0.1976, 0.1981, 0.2444, 0.2326, 0.2033, 0.2141, 0.1966,
0.2580, 0.2161, 0.2107, 0.2238, 0.2153, 0.2095, 0.1874,
0.2110, 0.1948, 0.1805, 0.2183, 0.2410, 0.2006, 0.1887,
0.2164, 0.2179, 0.2154, 0.1962, 0.2131, 0.1829, 0.2502,
0.2054, 0.2079, 0.2258, 0.2208, 0.2071, 0.1825, 0.2289,
0.2309, 0.2327, 0.2052, 0.2217, 0.2796, 0.1837, 0.1881,
0.2120, 0.2122, 0.2224, 0.2119, 0.2032, 0.2523, 0.3009,
0.1928, 0.1920, 0.2181, 0.2207, 0.2302, 0.2387, 0.2335,
0.1909, 0.1596, 0.1852, 0.1965, 0.2025, 0.1961, 0.2283,
0.1855, 0.2229, 0.2260, 0.2119, 0.2227, 0.2381, 0.1981,
0.2043, 0.1761, 0.2451, 0.2073, 0.2006, 0.2128, 0.2264,
0.2536, 0.2181, 0.1907, 0.2091, 0.2189, 0.1983, 0.2132,
0.2240, 0.2347, 0.1618, 0.1755, 0.1975, 0.2162, 0.2403,
0.2460, 0.2084, 0.1921, 0.2579, 0.2064, 0.1778, 0.1855,
0.1808, 0.1990, 0.2438, 0.1693, 0.2190, 0.2484, 0.2386,
0.1975, 0.2312, 0.1742, 0.2362, 0.2280, 0.2266, 0.2046,
0.2313, 0.2449, 0.1909, 0.2496, 0.2195, 0.1694, 0.2056,
0.2519, 0.2052, 0.2583, 0.1993, 0.1927, 0.2142, 0.1988,
0.2072, 0.2265, 0.1603, 0.2139, 0.2170, 0.1940, 0.2498,
0.1981, 0.1693, 0.1991, 0.1956, 0.1776, 0.2348, 0.2323,
0.2032, 0.2162, 0.2453, 0.2342, 0.2107, 0.2126, 0.2137,
0.2032, 0.2320, 0.2404, 0.2157, 0.2057, 0.1825, 0.2086,
0.2084, 0.2540, 0.2462, 0.1498, 0.2091, 0.1983, 0.2414,
0.1756], device='cuda:0')), ('layer4.1.bn2.bias', tensor([-0.1189, -0.1479, -0.0856, -0.1224, -0.1938, -0.1684, -0.0637,
-0.1032, -0.1072, -0.1162, -0.1292, -0.1759, -0.1509, -0.1312,
-0.1272, -0.1219, -0.1494, -0.1150, -0.1217, -0.1072, -0.1063,
-0.1308, -0.0865, -0.1324, -0.1705, -0.1196, -0.0586, -0.1192,
-0.1459, -0.1361, -0.0992, -0.0662, -0.1632, -0.0577, -0.1579,
-0.0796, -0.1192, -0.1050, -0.1017, -0.1539, -0.1534, -0.0863,
-0.0977, -0.0564, -0.1191, -0.0780, -0.0844, -0.1349, -0.1254,
-0.1476, -0.1117, -0.0777, 0.0425, -0.1330, -0.1465, -0.1373,
-0.1154, -0.1445, -0.1006, -0.2806, -0.1329, -0.1364, -0.1722,
-0.0949, -0.0135, -0.1355, -0.1422, -0.1664, -0.1112, -0.0518,
-0.1449, -0.0892, -0.1552, -0.1376, -0.0699, -0.2120, -0.0973,
-0.0242, -0.1322, -0.0740, -0.1921, -0.1017, -0.1564, -0.1202,
-0.0490, -0.1068, -0.1269, -0.1049, -0.1346, -0.1357, -0.1310,
-0.1704, -0.1467, -0.0764, -0.1455, -0.1008, -0.1234, -0.0895,
-0.1711, -0.1596, -0.0822, -0.0836, -0.0539, -0.0906, -0.0747,
-0.0126, -0.1221, -0.1593, -0.1948, -0.0421, -0.1108, -0.1071,
-0.1084, -0.0651, -0.1672, -0.0656, -0.1528, -0.1259, -0.1116,
-0.0770, -0.2188, -0.1074, -0.1425, -0.0493, -0.0976, -0.0596,
-0.1528, -0.0592, -0.1339, -0.1545, -0.1120, -0.1452, -0.1677,
-0.0652, -0.0927, -0.3386, -0.0783, -0.0912, -0.0545, -0.1454,
-0.0855, -0.1856, -0.1002, -0.1998, -0.1928, -0.0919, -0.1159,
-0.1016, -0.1273, -0.0696, -0.1820, -0.1310, -0.1699, -0.1200,
-0.0473, -0.2031, -0.1864, -0.0928, -0.1117, -0.1636, -0.0567,
-0.2695, -0.0781, -0.2245, -0.1768, -0.0860, -0.0970, -0.2098,
-0.0820, -0.1238, -0.1374, -0.0579, -0.1665, -0.0811, -0.0860,
-0.1434, -0.1570, -0.1080, -0.1648, -0.0697, -0.0570, -0.0737,
-0.1350, -0.1416, -0.1220, -0.0835, -0.1033, -0.1185, -0.1402,
-0.1128, -0.2154, -0.0889, -0.1191, -0.1345, -0.1425, -0.1006,
-0.0892, -0.1483, -0.1530, -0.1843, -0.0983, -0.1054, -0.0666,
-0.0926, -0.1845, -0.1749, -0.0680, -0.1200, -0.1293, -0.1993,
-0.1828, -0.1397, -0.1091, -0.1121, -0.1336, -0.1017, -0.1033,
-0.0542, -0.1028, -0.1423, -0.1070, -0.0546, -0.0664, -0.1555,
-0.1631, -0.0707, -0.0651, -0.1424, -0.1641, -0.1506, -0.1548,
-0.1511, -0.0472, -0.0514, -0.1399, -0.1263, -0.1048, -0.0939,
-0.1177, -0.2001, -0.1267, -0.0807, -0.1294, -0.2022, -0.0740,
-0.0932, -0.1397, -0.0898, -0.1524, -0.1606, -0.0582, -0.1467,
-0.0999, -0.0872, 0.0069, -0.1458, -0.0801, -0.1334, -0.0943,
-0.1327, -0.0924, -0.1014, -0.0587, -0.0916, -0.0744, -0.0958,
-0.0181, -0.0931, -0.1618, -0.1269, -0.1757, 0.0006, -0.1169,
-0.1453, -0.1795, -0.1743, -0.1662, -0.0411, -0.1156, -0.0171,
-0.1857, -0.1683, -0.1041, -0.1351, -0.1635, -0.1486, -0.1680,
-0.1276, -0.0680, -0.1430, -0.1310, -0.2915, -0.1773, -0.1072,
-0.1389, -0.2081, -0.1270, -0.0799, -0.1273, -0.1426, -0.1397,
-0.1273, -0.0559, -0.0872, -0.0873, -0.1202, -0.0877, -0.0640,
-0.1409, -0.1445, -0.1100, -0.1509, -0.0758, -0.1131, -0.1093,
0.0805, -0.1692, -0.1363, -0.1765, -0.1192, -0.1138, -0.2116,
-0.1297, -0.0901, -0.1053, -0.1055, -0.2111, -0.1460, -0.1426,
-0.0526, -0.1220, -0.1547, -0.1350, -0.1527, -0.0388, -0.0462,
-0.0446, -0.1157, -0.1437, -0.0409, -0.0946, -0.0769, -0.1120,
-0.1162, -0.0431, -0.0736, -0.1349, -0.0736, -0.1052, -0.1522,
-0.1467, -0.1317, -0.1682, -0.1269, -0.0424, -0.1161, -0.1573,
-0.1290, -0.1214, -0.0934, -0.2118, -0.1547, -0.1011, -0.1449,
-0.1261, -0.1626, -0.0858, -0.1778, -0.1853, -0.1311, -0.1175,
-0.1343, -0.2016, -0.1605, -0.0673, -0.1403, -0.0624, -0.1535,
-0.1787, -0.1105, -0.1674, -0.1756, -0.1189, -0.0923, -0.1046,
-0.1358, -0.1293, -0.0818, -0.0801, -0.1398, -0.0364, -0.1776,
-0.1591, -0.1396, -0.0833, -0.1685, -0.1528, -0.1099, -0.3568,
-0.1117, -0.0720, -0.1625, -0.1109, -0.1734, -0.1132, -0.1038,
-0.0527, -0.0584, -0.1460, -0.1880, -0.1275, -0.0860, -0.1231,
-0.1516, -0.0895, -0.1122, -0.1295, -0.1469, -0.1274, -0.1139,
-0.1749, -0.0656, -0.1275, -0.1179, -0.0536, -0.1574, -0.1396,
-0.1407, -0.1024, -0.1194, -0.0588, -0.1184, -0.1091, -0.1108,
-0.1244, -0.1383, -0.0408, -0.0381, -0.0658, -0.1089, -0.1964,
-0.1383, -0.1129, -0.1108, -0.1645, -0.1954, -0.0578, -0.0432,
-0.0985, -0.1146, -0.0578, -0.1195, -0.1079, -0.1289, -0.1115,
-0.1305, -0.1669, -0.1982, -0.1299, -0.1371, -0.1800, -0.0112,
-0.1111, -0.1518, -0.0571, -0.1401, -0.1064, -0.0788, -0.1134,
-0.1495, -0.1327, -0.1025, -0.1378, -0.0733, -0.0895, -0.1002,
-0.1370, -0.2019, -0.1350, -0.0642, -0.1117, -0.1591, -0.0846,
-0.1379, -0.1465, -0.0808, -0.0821, -0.1765, -0.1097, -0.1215,
-0.1092, 0.0387, -0.1725, -0.1062, -0.1151, -0.0987, -0.1511,
-0.1891, -0.1301, -0.1304, -0.1479, -0.1036, -0.1424, -0.1432,
-0.0631, -0.1372, -0.1780, -0.0260, -0.1124, 0.0163, -0.0710,
-0.0830], device='cuda:0')), ('layer4.1.bn2.running_mean', tensor([-0.1046, -0.1131, -0.1054, -0.1396, -0.0614, -0.1146, -0.1392,
-0.1108, -0.0467, -0.1512, -0.0558, -0.1296, -0.0775, -0.0963,
-0.0571, -0.1152, -0.0905, -0.0317, -0.0456, -0.1036, -0.0960,
-0.0724, -0.1672, -0.1026, -0.1256, -0.0997, -0.0729, -0.1994,
-0.0949, -0.0982, -0.1415, -0.0667, -0.0442, -0.0471, -0.0473,
-0.0558, -0.1038, -0.1190, -0.1163, -0.0494, -0.1231, -0.2104,
-0.1042, -0.0448, -0.0993, -0.0845, -0.0166, -0.1138, -0.2062,
-0.0933, -0.0815, -0.1126, 0.4061, 0.0006, -0.0556, -0.1253,
-0.0205, -0.0755, -0.0610, -0.0947, -0.1818, -0.0974, -0.0439,
0.1409, -0.0387, -0.1117, -0.1651, -0.0384, -0.1108, -0.1019,
-0.1871, -0.1775, -0.0813, -0.0645, -0.0631, -0.0888, -0.0852,
-0.1090, -0.1249, -0.0954, -0.1034, -0.0422, -0.1005, -0.1065,
-0.2052, -0.1223, -0.0596, -0.0449, -0.1371, -0.1054, -0.1229,
-0.1354, -0.0903, -0.0359, -0.0882, -0.0831, -0.0741, -0.1458,
-0.0650, -0.1167, -0.1692, -0.1972, -0.1154, -0.1248, -0.1386,
-0.1366, -0.1155, -0.1126, -0.0747, -0.0236, -0.1452, -0.0773,
-0.2051, -0.1127, -0.1111, -0.1005, -0.1296, -0.2153, -0.1608,
-0.0806, -0.0623, -0.0852, -0.0915, -0.0675, -0.0808, -0.0629,
-0.0718, -0.0401, -0.1409, -0.1024, -0.0918, -0.1168, -0.0495,
-0.0538, -0.0146, -0.2991, -0.0654, -0.0542, -0.0609, -0.0932,
-0.0239, -0.0900, -0.1216, -0.0925, -0.0776, -0.1285, -0.1086,
-0.2390, -0.0773, -0.1664, -0.1360, -0.2691, -0.1434, -0.0566,
-0.0782, -0.1409, -0.1330, -0.0935, -0.1521, -0.0983, 0.0109,
-0.1246, -0.0615, -0.1096, -0.0952, -0.2090, -0.1163, -0.0784,
-0.1544, -0.0790, -0.0628, -0.1205, -0.1298, -0.1187, -0.1886,
-0.1293, -0.0725, -0.0880, -0.0608, -0.1180, -0.0860, -0.1394,
-0.1073, -0.0856, -0.1090, -0.0741, -0.0925, -0.1069, -0.1871,
-0.1518, -0.1053, -0.1526, -0.1440, -0.1484, 0.0073, -0.0841,
-0.0157, -0.1063, -0.0659, -0.1353, -0.2034, -0.0024, -0.0410,
-0.1106, -0.0954, -0.0810, -0.0097, -0.0623, -0.1219, -0.1000,
-0.0902, -0.1142, -0.2189, -0.1245, -0.0845, -0.0856, -0.1847,
-0.0824, -0.1005, -0.0674, -0.1382, -0.0343, -0.0833, 0.0413,
-0.0729, -0.0326, -0.1456, -0.0901, -0.0209, -0.0925, -0.1979,
-0.0478, -0.1052, -0.1770, -0.1362, -0.0878, -0.2544, -0.1001,
-0.1440, -0.1072, -0.0753, -0.1210, -0.0959, -0.0998, -0.0630,
-0.0647, -0.1155, -0.1503, -0.0958, -0.0942, -0.1480, -0.0436,
-0.0711, -0.1486, -0.0731, -0.1146, -0.0874, -0.0484, -0.1952,
-0.2165, -0.0164, -0.0591, -0.0619, -0.0627, -0.1274, -0.1122,
-0.1827, -0.0053, -0.1006, -0.0508, -0.1846, -0.1991, -0.1257,
-0.1068, -0.0839, -0.1381, -0.0775, -0.1195, -0.0824, -0.1479,
-0.1763, -0.0557, -0.0563, -0.0367, -0.1369, -0.0762, -0.1436,
-0.1201, -0.1446, -0.1595, -0.0212, -0.1040, -0.1248, -0.0484,
-0.0621, -0.1691, -0.0619, -0.0162, -0.0791, 0.0146, -0.1189,
-0.0909, -0.0915, -0.0863, -0.0995, -0.1614, -0.2426, -0.0670,
-0.0860, -0.0727, -0.0452, -0.1422, -0.0759, -0.1322, -0.0737,
-0.0050, -0.1238, -0.0824, -0.1118, -0.0832, -0.1436, -0.0899,
-0.1042, -0.0877, -0.1838, -0.0855, -0.1281, -0.0818, -0.1063,
-0.0618, -0.1222, -0.1308, -0.1359, -0.0630, -0.0705, -0.1351,
-0.1025, -0.0577, -0.0936, -0.1666, -0.0894, -0.1695, -0.1694,
-0.0682, -0.1841, -0.0661, -0.0943, -0.0830, -0.0427, -0.1411,
-0.0986, -0.1306, -0.1072, -0.1840, -0.0500, -0.1277, -0.1345,
-0.1281, -0.1724, -0.1759, -0.1384, -0.1297, -0.0477, -0.1493,
-0.1302, -0.0946, -0.1020, -0.1800, -0.0711, -0.0305, -0.1285,
-0.1967, -0.0688, -0.1000, -0.0564, -0.0590, -0.0937, -0.2190,
-0.0476, -0.1224, -0.0703, -0.0746, -0.1216, -0.0379, -0.1205,
-0.1684, -0.0533, -0.1349, -0.1053, -0.0620, -0.0857, -0.1234,
-0.1758, -0.0810, -0.1230, -0.1304, -0.1180, -0.0979, -0.4198,
-0.1104, -0.0852, -0.1226, -0.1675, -0.1128, -0.0906, -0.0406,
-0.0846, -0.2491, -0.1032, -0.1159, -0.1000, -0.1344, -0.0329,
-0.1319, -0.0283, -0.0817, -0.0488, -0.1388, -0.1662, -0.1022,
-0.0142, -0.1222, -0.1143, -0.0765, -0.1119, -0.1593, -0.0970,
-0.1118, -0.0872, -0.1545, -0.0356, -0.1023, -0.1336, -0.1283,
-0.1200, -0.0262, -0.1345, -0.0136, -0.1787, -0.0005, -0.0678,
-0.1541, -0.1778, -0.0237, -0.1457, -0.1049, -0.0579, -0.2134,
-0.0921, -0.1261, -0.1020, -0.1585, -0.0962, -0.1114, -0.0470,
-0.0668, -0.1572, -0.0911, -0.0693, -0.0283, -0.1328, -0.0646,
-0.0873, -0.1687, -0.1178, -0.1384, -0.1344, -0.0439, -0.1000,
0.4463, -0.1062, -0.0858, -0.0806, -0.0511, -0.0589, -0.0557,
-0.1070, -0.0995, -0.0940, -0.0217, -0.0438, -0.0783, -0.0972,
-0.0956, -0.1486, -0.1182, -0.0599, -0.1295, -0.1232, -0.0487,
-0.1322, -0.1236, -0.0833, -0.0978, -0.0918, -0.1587, -0.1975,
-0.0911, -0.1145, -0.0428, -0.0165, -0.2068, -0.1503, -0.1082,
-0.0462, -0.0619, -0.1996, -0.0526, -0.1325, -0.1517, -0.1237,
-0.1239], device='cuda:0')), ('layer4.1.bn2.running_var', tensor(1.00000e-02 *
[ 3.9833, 1.0967, 1.3495, 1.7448, 0.8461, 1.4367, 1.9028,
1.5660, 1.0297, 1.3851, 1.2571, 1.5072, 1.0962, 1.1774,
1.0699, 2.2131, 1.4564, 1.1743, 1.3488, 1.5087, 1.1875,
1.0700, 1.9780, 1.5113, 1.5281, 1.2812, 1.2925, 2.7793,
1.2972, 1.5629, 2.1963, 1.3221, 1.3329, 0.9373, 1.0902,
1.3658, 1.1464, 1.6158, 1.6031, 1.3325, 1.5170, 2.5412,
1.4953, 1.5582, 1.1624, 1.4205, 1.1790, 1.4191, 2.0786,
1.5523, 1.2365, 1.5099, 4.1409, 1.0290, 1.1758, 1.2022,
1.6383, 1.6772, 1.1789, 0.8937, 2.4244, 2.2013, 1.0784,
1.4335, 1.0242, 1.9825, 2.1914, 0.8584, 1.0454, 1.9206,
1.7562, 1.3747, 1.2861, 1.3596, 1.5054, 1.2412, 1.0121,
1.8820, 1.2598, 2.3116, 1.1094, 1.6330, 1.2327, 1.2038,
3.6857, 2.1789, 1.4112, 1.4074, 1.4548, 1.2612, 1.1021,
1.7085, 1.4778, 2.1732, 1.4721, 1.4277, 1.1455, 2.9518,
1.2423, 1.1863, 1.4742, 1.9302, 2.0700, 1.6044, 1.2726,
1.6789, 1.4957, 1.2204, 1.1908, 1.7827, 2.0010, 1.0137,
2.8950, 1.3191, 1.3908, 1.5170, 1.4664, 1.9138, 1.4817,
1.7047, 0.8164, 1.2085, 0.9528, 1.7028, 2.2372, 1.1212,
0.9993, 1.1437, 1.5680, 1.0110, 1.0410, 1.0550, 1.6938,
1.0708, 1.1229, 4.2579, 1.8829, 1.1843, 1.5400, 1.2722,
1.2015, 1.1128, 1.4218, 1.0007, 0.9500, 1.2824, 1.3944,
3.1350, 1.2281, 1.5473, 1.4851, 2.0506, 1.0287, 0.9524,
1.8829, 1.5150, 0.9982, 1.2973, 1.4386, 1.2814, 1.1476,
1.2023, 1.5441, 1.0609, 1.5207, 2.6408, 1.6424, 1.1148,
3.0852, 1.3053, 1.1069, 1.8300, 1.5561, 1.4212, 2.5174,
1.4941, 1.0146, 1.9765, 1.1933, 1.4072, 1.5220, 1.7243,
1.6738, 1.0932, 1.4956, 1.1706, 1.1994, 1.2256, 2.1317,
1.8615, 1.1963, 1.7880, 1.0641, 1.8015, 1.2019, 1.1625,
1.1120, 1.2259, 1.1645, 1.3843, 2.2553, 1.3831, 2.0168,
1.3497, 1.0351, 1.1245, 1.3822, 1.3144, 1.8431, 1.0802,
1.0608, 0.9519, 1.9493, 1.0265, 1.4527, 1.4439, 2.0486,
1.3701, 1.5223, 1.0259, 1.8650, 0.8897, 1.4631, 2.2092,
0.9111, 1.4967, 1.8583, 1.0676, 0.9149, 1.2299, 1.9417,
1.0887, 2.3940, 2.2801, 1.5306, 1.2544, 3.2695, 1.1629,
1.7486, 1.5842, 0.9544, 1.3385, 1.3177, 1.1421, 1.1491,
1.4834, 1.2959, 1.9014, 1.3395, 1.3647, 2.0478, 1.8487,
2.0041, 1.4874, 1.3498, 1.3354, 1.4893, 1.3639, 1.9255,
2.1928, 1.1271, 1.0761, 1.3991, 1.1737, 1.7652, 1.3473,
3.4247, 1.3453, 0.9614, 1.2620, 2.5936, 3.5633, 1.4272,
1.2545, 1.4737, 1.6250, 1.0717, 1.9125, 1.2447, 2.3620,
1.3000, 0.7783, 1.0608, 1.0900, 1.1350, 1.5744, 1.7598,
1.6093, 2.3986, 1.2081, 1.1661, 2.5847, 1.3631, 1.0457,
1.3835, 2.0767, 1.4888, 0.9273, 1.0622, 1.1831, 1.5226,
1.2128, 1.2606, 1.6345, 1.4196, 1.6327, 2.4114, 2.3322,
1.0664, 1.1416, 1.2247, 1.5702, 1.8861, 1.5743, 1.2760,
1.7822, 1.5788, 1.7427, 1.5710, 0.9847, 2.3834, 1.1829,
1.4697, 1.4836, 1.6157, 1.1226, 1.2501, 1.6977, 1.2562,
1.2318, 1.4403, 1.4302, 1.6642, 0.8282, 2.1291, 3.0687,
2.2213, 0.9726, 1.2519, 2.7147, 1.3119, 2.2896, 1.8852,
1.6011, 2.4991, 1.7240, 1.3764, 2.3374, 1.2977, 1.4681,
1.0511, 1.6691, 1.4802, 2.5217, 1.3635, 1.6618, 1.2790,
1.5317, 2.5406, 2.2201, 1.5252, 1.4449, 0.9012, 2.1587,
1.4751, 1.2780, 1.3923, 1.7329, 1.3008, 1.2647, 1.4007,
2.4482, 1.0806, 1.1778, 1.3758, 1.3914, 1.0997, 2.4231,
1.2075, 1.2696, 1.2408, 1.2633, 1.5785, 1.3592, 1.6801,
1.6335, 1.2346, 1.2857, 1.6452, 1.4805, 1.7528, 1.1604,
1.4748, 1.2720, 1.7785, 0.8882, 1.3612, 1.7200, 5.1882,
1.6017, 1.3853, 1.1907, 1.5093, 1.2678, 1.4247, 1.5974,
1.2421, 2.8341, 0.9115, 1.0116, 1.5232, 1.2129, 0.9770,
1.4654, 1.2296, 2.3084, 1.3778, 1.4532, 1.8162, 1.2942,
0.8807, 2.4699, 2.3350, 1.4394, 1.8595, 1.9409, 1.2726,
2.0752, 1.5631, 1.5242, 1.5383, 1.1046, 2.1937, 1.7928,
1.3004, 1.1316, 1.2788, 0.9694, 2.4622, 1.1826, 1.2911,
1.3002, 2.1148, 1.3447, 1.3793, 1.3223, 1.2923, 2.7524,
1.0003, 1.5482, 2.2608, 1.8505, 1.3230, 1.5412, 1.5051,
1.1631, 1.7029, 1.0060, 1.3208, 1.3424, 1.7719, 2.0225,
2.0263, 2.4655, 1.9472, 1.5902, 1.4574, 1.1394, 1.5684,
3.0373, 1.3149, 1.8318, 1.1278, 1.0818, 1.6710, 1.1538,
1.7517, 1.4704, 1.2292, 1.3644, 1.1347, 1.0023, 1.8464,
1.6704, 0.9799, 1.3835, 1.3031, 1.0933, 1.6835, 1.0572,
1.6660, 2.6578, 1.0699, 1.7643, 1.4479, 1.6104, 3.0193,
1.0477, 1.5171, 0.9995, 1.2940, 2.0889, 1.3446, 1.2474,
1.1854, 1.0999, 2.8599, 1.3337, 1.2899, 1.8405, 1.7655,
1.2587], device='cuda:0')), ('layer4.1.conv3.weight', tensor([[[[ 3.3387e-03]],
[[-5.6135e-03]],
[[ 6.8908e-03]],
...,
[[-1.1271e-02]],
[[ 2.4669e-03]],
[[-2.5670e-03]]],
[[[ 8.3620e-03]],
[[-4.8913e-02]],
[[ 3.0107e-02]],
...,
[[-2.1985e-02]],
[[-1.3234e-02]],
[[ 2.1861e-02]]],
[[[ 2.1996e-03]],
[[-3.8735e-03]],
[[-2.3089e-02]],
...,
[[ 5.2568e-03]],
[[-1.2967e-02]],
[[-9.4131e-04]]],
...,
[[[ 2.2241e-02]],
[[-2.1514e-02]],
[[ 5.3832e-03]],
...,
[[-6.3454e-03]],
[[ 3.4191e-02]],
[[-7.1475e-04]]],
[[[-5.2657e-03]],
[[ 2.3503e-02]],
[[ 5.4598e-03]],
...,
[[ 3.1585e-03]],
[[ 1.0974e-02]],
[[ 2.5095e-03]]],
[[[ 6.3988e-03]],
[[-4.3090e-04]],
[[-1.6043e-02]],
...,
[[-7.1117e-04]],
[[ 5.5824e-03]],
[[ 5.9582e-03]]]], device='cuda:0')), ('layer4.1.bn3.weight', tensor([ 0.2978, 0.5030, 0.3389, ..., 0.2588, 0.3209, 0.2572], device='cuda:0')), ('layer4.1.bn3.bias', tensor([-0.0863, -0.0775, -0.0690, ..., -0.0764, -0.0860, -0.0737], device='cuda:0')), ('layer4.1.bn3.running_mean', tensor([ 2.4509e-03, 4.5755e-03, -9.9135e-03, ..., -1.7449e-03,
-1.3096e-03, -2.0193e-03], device='cuda:0')), ('layer4.1.bn3.running_var', tensor(1.00000e-03 *
[ 0.7308, 1.0537, 0.9476, ..., 0.5121, 0.6400, 0.4531], device='cuda:0')), ('layer4.2.conv1.weight', tensor([[[[ 1.3010e-02]],
[[ 1.3665e-02]],
[[ 2.5826e-03]],
...,
[[ 1.6641e-02]],
[[ 1.4051e-02]],
[[ 5.2212e-03]]],
[[[-2.4446e-02]],
[[ 6.9436e-03]],
[[-1.8915e-02]],
...,
[[-2.8402e-02]],
[[-6.1963e-03]],
[[ 1.9334e-03]]],
[[[-5.1116e-03]],
[[-6.8200e-03]],
[[-1.5854e-02]],
...,
[[ 7.6137e-03]],
[[-7.1213e-03]],
[[-2.5644e-04]]],
...,
[[[-9.4383e-03]],
[[ 1.8722e-02]],
[[-1.6677e-02]],
...,
[[ 7.1453e-03]],
[[ 1.8863e-02]],
[[ 1.8104e-02]]],
[[[-1.8470e-02]],
[[-6.6378e-03]],
[[-1.6084e-03]],
...,
[[-8.0362e-04]],
[[ 6.3490e-03]],
[[ 3.8941e-02]]],
[[[ 5.7269e-03]],
[[ 3.7425e-03]],
[[-3.1622e-02]],
...,
[[-3.1648e-02]],
[[-3.0168e-02]],
[[ 8.1424e-03]]]], device='cuda:0')), ('layer4.2.bn1.weight', tensor([ 0.2363, 0.1918, 0.2071, 0.2235, 0.2186, 0.1923, 0.2560,
0.2263, 0.1935, 0.1766, 0.2011, 0.2089, 0.2782, 0.2550,
0.2280, 0.2464, 0.1968, 0.2230, 0.2142, 0.2278, 0.1639,
0.2074, 0.1856, 0.1902, 0.2044, 0.2336, 0.1845, 0.1963,
0.2244, 0.2136, 0.1851, 0.2673, 0.2160, 0.2225, 0.2685,
0.1534, 0.2372, 0.2304, 0.2167, 0.1370, 0.2561, 0.2132,
0.2224, 0.1869, 0.2012, 0.2075, 0.2379, 0.2253, 0.2040,
0.2682, 0.1928, 0.1540, 0.2018, 0.2770, 0.2257, 0.2016,
0.2606, 0.2504, 0.1955, 0.2128, 0.2094, 0.2368, 0.2380,
0.1286, 0.2374, 0.2476, 0.1465, 0.1846, 0.1682, 0.2364,
0.2348, 0.2574, 0.1238, 0.2298, 0.1267, 0.2160, 0.2360,
0.1822, 0.2508, 0.2116, 0.1871, 0.2345, 0.1634, 0.1452,
0.1998, 0.2324, 0.1776, 0.2420, 0.2406, 0.2404, 0.2029,
0.2653, 0.2667, 0.2347, 0.2619, 0.2600, 0.2077, 0.2387,
0.2491, 0.2421, 0.2042, 0.2178, 0.2435, 0.2532, 0.2203,
0.2395, 0.2125, 0.2857, 0.1489, 0.2433, 0.2422, 0.2243,
0.2439, 0.2254, 0.1854, 0.2201, 0.2281, 0.2272, 0.2467,
0.2210, 0.2072, 0.2303, 0.2148, 0.1493, 0.1717, 0.1870,
0.2373, 0.1579, 0.2424, 0.2215, 0.1964, 0.2655, 0.2448,
0.1864, 0.1771, 0.2798, 0.2227, 0.2097, 0.2727, 0.1398,
0.2197, 0.1800, 0.2256, 0.2162, 0.2064, 0.2281, 0.2355,
0.2178, 0.2223, 0.2373, 0.1714, 0.2493, 0.2442, 0.2235,
0.2107, 0.2213, 0.2260, 0.2531, 0.2020, 0.2384, 0.1991,
0.0966, 0.2210, 0.1988, 0.2425, 0.2320, 0.2086, 0.1937,
0.2000, 0.2167, 0.2838, 0.2320, 0.2054, 0.2079, 0.2212,
0.2151, 0.1614, 0.2019, 0.2360, 0.2216, 0.1486, 0.2399,
0.2455, 0.2257, 0.1789, 0.2180, 0.2599, 0.2497, 0.2249,
0.1978, 0.2477, 0.2337, 0.1111, 0.2835, 0.2175, 0.2425,
0.2361, 0.2601, 0.2429, 0.2326, 0.2462, 0.1726, 0.2207,
0.2672, 0.1534, 0.2606, 0.2046, 0.2206, 0.1914, 0.2524,
0.2494, 0.2134, 0.2402, 0.2116, 0.1995, 0.2532, 0.1658,
0.1871, 0.2709, 0.1785, 0.2221, 0.2565, 0.1437, 0.1296,
0.2480, 0.2227, 0.1212, 0.1823, 0.2075, 0.1798, 0.2305,
0.2462, 0.2540, 0.2553, 0.2185, 0.2483, 0.2182, 0.2007,
0.2162, 0.2132, 0.2334, 0.2057, 0.2158, 0.1984, 0.2272,
0.1910, 0.1922, 0.2136, 0.2358, 0.2786, 0.2225, 0.2926,
0.1711, 0.2500, 0.1900, 0.2469, 0.2342, 0.2215, 0.2474,
0.2558, 0.2262, 0.2239, 0.2486, 0.1982, 0.2116, 0.2225,
0.2141, 0.2525, 0.2361, 0.2295, 0.2116, 0.2235, 0.1797,
0.2249, 0.2174, 0.1615, 0.3090, 0.1643, 0.2008, 0.2009,
0.2174, 0.2213, 0.1506, 0.1227, 0.2105, 0.2093, 0.2419,
0.2357, 0.2135, 0.1787, 0.2459, 0.2279, 0.2549, 0.2376,
0.2540, 0.1499, 0.2399, 0.2331, 0.2260, 0.2249, 0.1863,
0.2756, 0.0871, 0.2291, 0.2613, 0.2707, 0.2190, 0.2290,
0.1418, 0.2432, 0.1222, 0.2472, 0.1146, 0.2083, 0.2326,
0.1740, 0.2342, 0.2354, 0.1750, 0.2302, 0.2471, 0.1994,
0.1813, 0.1790, 0.2377, 0.2335, 0.1762, 0.2231, 0.2151,
0.2463, 0.2527, 0.2514, 0.2342, 0.1952, 0.1949, 0.2601,
0.2138, 0.1706, 0.2457, 0.0957, 0.1571, 0.2431, 0.1934,
0.2459, 0.2911, 0.2377, 0.1448, 0.2688, 0.2244, 0.2366,
0.2076, 0.1722, 0.2195, 0.2341, 0.2574, 0.1316, 0.2157,
0.2282, 0.2323, 0.2068, 0.2501, 0.2197, 0.2536, 0.1752,
0.2501, 0.2196, 0.2019, 0.2385, 0.1601, 0.2132, 0.5445,
0.2051, 0.2071, 0.2119, 0.2490, 0.1849, 0.2543, 0.2093,
0.2445, 0.1333, 0.2273, 0.1684, 0.2122, 0.2409, 0.1936,
0.1500, 0.2113, 0.2412, 0.2118, 0.2013, 0.2277, 0.2548,
0.2176, 0.1448, 0.1893, 0.1650, 0.2161, 0.1357, 0.2457,
0.2390, 0.2206, 0.2725, 0.2062, 0.2582, 0.2884, 0.2083,
0.2545, 0.2450, 0.2185, 0.2025, 0.2041, 0.2013, 0.1797,
0.1867, 0.1951, 0.2434, 0.0941, 0.2397, 0.2098, 0.2167,
0.1716, 0.1981, 0.2394, 0.1984, 0.2161, 0.1639, 0.2208,
0.2349, 0.2182, 0.1353, 0.2473, 0.2168, 0.1769, 0.2122,
0.2350, 0.3040, 0.2199, 0.2457, 0.2413, 0.2202, 0.2317,
0.2207, 0.2504, 0.2948, 0.1854, 0.2407, 0.2658, 0.2647,
0.1919, 0.1933, 0.1681, 0.2182, 0.1319, 0.2085, 0.2071,
0.2643, 0.2256, 0.2415, 0.1903, 0.2541, 0.2715, 0.2393,
0.2333, 0.2338, 0.2639, 0.2009, 0.2510, 0.1645, 0.2083,
0.1873, 0.1947, 0.2387, 0.2463, 0.2361, 0.1996, 0.2173,
0.1981, 0.2116, 0.2129, 0.2554, 0.2576, 0.2350, 0.2334,
0.2264, 0.2557, 0.2212, 0.1522, 0.2565, 0.1886, 0.1858,
0.1786, 0.2265, 0.1823, 0.2339, 0.2604, 0.1193, 0.2428,
0.2359, 0.2170, 0.2252, 0.1751, 0.2314, 0.1616, 0.2606,
0.2979, 0.2539, 0.1939, 0.2340, 0.2356, 0.2163, 0.2108,
0.2119], device='cuda:0')), ('layer4.2.bn1.bias', tensor([-0.1444, -0.1619, -0.1992, -0.2047, -0.1642, -0.2173, -0.2018,
-0.1990, -0.1170, -0.1323, -0.1071, -0.1439, -0.2062, -0.2187,
-0.1902, -0.1875, -0.0569, -0.1687, -0.1655, -0.2083, -0.0706,
-0.1985, -0.0996, -0.1845, -0.1888, -0.1338, -0.2012, -0.1915,
-0.1603, -0.1566, -0.1610, -0.2606, -0.2012, -0.2021, -0.1664,
-0.0834, -0.2043, -0.1779, -0.1402, -0.0363, -0.1371, -0.1859,
-0.1817, -0.1276, -0.1576, -0.1787, -0.2032, -0.1502, -0.2092,
-0.2784, -0.1749, -0.1394, -0.1377, -0.2232, -0.2127, -0.1439,
-0.2202, -0.2184, -0.1972, -0.1261, -0.1404, -0.2149, -0.1659,
-0.0269, -0.2002, -0.1917, -0.0901, -0.0842, -0.1154, -0.2141,
-0.2279, -0.2240, -0.0635, -0.2048, -0.0106, -0.1520, -0.1482,
-0.2001, -0.2400, -0.1969, -0.0859, -0.1712, -0.1093, -0.0491,
-0.1466, -0.2237, -0.1112, -0.2172, -0.1786, -0.2171, -0.1333,
-0.2795, -0.2499, -0.1910, -0.2175, -0.2097, -0.2030, -0.2634,
-0.2525, -0.1919, -0.2023, -0.1312, -0.1877, -0.2326, -0.1378,
-0.1579, -0.1544, -0.2165, -0.0958, -0.2585, -0.2257, -0.1769,
-0.1704, -0.1580, -0.1170, -0.1577, -0.1706, -0.1916, -0.2424,
-0.2003, -0.1719, -0.1771, -0.1815, -0.0658, -0.0797, -0.1551,
-0.2674, -0.0588, -0.2070, -0.2122, -0.1491, -0.1948, -0.2781,
-0.1708, -0.1402, -0.2296, -0.2066, -0.1845, -0.2528, -0.0620,
-0.1993, -0.1547, -0.1895, -0.1333, -0.1864, -0.1948, -0.1989,
-0.1654, -0.2095, -0.2730, -0.1324, -0.2355, -0.1881, -0.1554,
-0.1783, -0.1738, -0.1663, -0.2224, -0.1481, -0.2350, -0.1683,
0.0542, -0.1542, -0.2063, -0.2199, -0.1800, -0.0847, -0.2233,
-0.1993, -0.1862, -0.1952, -0.2349, -0.1559, -0.1648, -0.2462,
-0.1696, -0.0680, -0.1735, -0.1750, -0.1898, -0.1156, -0.1661,
-0.2120, -0.2231, -0.1274, -0.1086, -0.1827, -0.1488, -0.2290,
-0.1150, -0.1434, -0.2929, 0.0543, -0.2380, -0.2198, -0.1984,
-0.1426, -0.1606, -0.1802, -0.1898, -0.2079, -0.1601, -0.1515,
-0.2254, -0.0944, -0.2475, -0.1409, -0.1740, -0.0681, -0.2624,
-0.2036, -0.1423, -0.1745, -0.1738, -0.1690, -0.2206, -0.1785,
-0.2014, -0.2286, -0.1676, -0.1850, -0.2235, -0.0374, -0.0355,
-0.1457, -0.2089, -0.0804, -0.1779, -0.1390, -0.1619, -0.0921,
-0.1914, -0.1670, -0.1594, -0.1458, -0.2620, -0.1652, -0.1704,
-0.1613, -0.2273, -0.2257, -0.1385, -0.1779, -0.1360, -0.1448,
-0.1689, -0.2093, -0.2222, -0.1177, -0.2339, -0.1240, -0.3008,
-0.1693, -0.2125, -0.1943, -0.2106, -0.1588, -0.2341, -0.2223,
-0.1793, -0.2264, -0.1380, -0.2928, -0.0897, -0.1591, -0.2441,
-0.2030, -0.2567, -0.1156, -0.2225, -0.1652, -0.1600, -0.1571,
-0.1732, -0.2564, -0.1291, -0.1576, -0.0492, -0.2007, -0.1773,
-0.2090, -0.1302, -0.0972, -0.0025, -0.1150, -0.2350, -0.1151,
-0.2140, -0.1659, -0.1569, -0.2631, -0.1735, -0.1714, -0.2675,
-0.2278, -0.0817, -0.2114, -0.1833, -0.1588, -0.1440, -0.1103,
-0.2102, 0.0848, -0.1407, -0.2088, -0.1969, -0.1755, -0.2337,
-0.0188, -0.2418, 0.0585, -0.2156, 0.0454, -0.1859, -0.1629,
-0.1472, -0.1884, -0.2162, -0.1193, -0.1440, -0.2434, -0.1883,
-0.1631, -0.1703, -0.2046, -0.2287, -0.1444, -0.1955, -0.2399,
-0.2074, -0.2451, -0.1874, -0.2054, -0.1774, -0.2112, -0.2342,
-0.1518, -0.1511, -0.2534, 0.0997, -0.1392, -0.1779, -0.1363,
-0.1695, -0.2144, -0.1720, -0.0552, -0.2556, -0.2077, -0.1419,
-0.0872, -0.0899, -0.1610, -0.1937, -0.1985, -0.1108, -0.1942,
-0.1654, -0.1779, -0.1879, -0.2622, -0.1601, -0.1718, -0.1183,
-0.2475, -0.1918, -0.2021, -0.1667, -0.1558, -0.2332, -0.2980,
-0.1731, -0.1841, -0.1175, -0.1874, -0.1771, -0.2104, -0.1519,
-0.1320, -0.1132, -0.2025, -0.1058, -0.2317, -0.2141, -0.1792,
0.0254, -0.1937, -0.1820, -0.1723, -0.1753, -0.2229, -0.2171,
-0.1582, -0.0248, -0.1439, -0.1000, -0.1971, -0.1417, -0.2150,
-0.2555, -0.1624, -0.2250, -0.2159, -0.2598, -0.2474, -0.1970,
-0.1826, -0.1961, -0.1826, -0.1864, -0.1484, -0.1681, -0.1124,
-0.0861, -0.1876, -0.1972, 0.0309, -0.1782, -0.1204, -0.1880,
-0.0636, -0.1943, -0.2019, -0.1549, -0.1898, -0.1118, -0.1888,
-0.1662, -0.2588, -0.0974, -0.1607, -0.1077, -0.1885, -0.1835,
-0.2055, -0.2521, -0.2145, -0.1400, -0.2073, -0.0913, -0.2314,
-0.1624, -0.2199, -0.2229, -0.1243, -0.1538, -0.1650, -0.2730,
-0.1517, -0.1650, -0.1312, -0.2062, 0.0310, -0.2337, -0.1821,
-0.1653, -0.1823, -0.1637, -0.1265, -0.2064, -0.2294, -0.1877,
-0.1458, -0.1671, -0.2625, -0.1062, -0.2809, -0.1051, -0.2054,
-0.0831, -0.1214, -0.1166, -0.1262, -0.2134, -0.1761, -0.1655,
-0.1557, -0.1365, -0.2408, -0.2929, -0.1812, -0.1897, -0.1874,
-0.2324, -0.2316, -0.2161, -0.0776, -0.1334, -0.1072, -0.1428,
-0.1387, -0.1922, -0.1591, -0.2086, -0.2593, 0.0611, -0.2169,
-0.2147, -0.1987, -0.0726, -0.1559, -0.2132, -0.1104, -0.1641,
-0.3384, -0.2620, -0.1259, -0.1451, -0.1509, -0.1360, -0.1944,
-0.2236], device='cuda:0')), ('layer4.2.bn1.running_mean', tensor([-0.4346, -0.7203, -0.3897, -0.4644, -0.3208, -0.5077, -0.0772,
-0.3425, -0.3170, -0.2733, -0.6527, -0.4747, -0.3648, -0.4468,
-0.3581, -0.3320, 0.9340, -0.2337, -0.3631, -0.2625, -0.1415,
-0.0997, -0.2469, -0.1822, -0.6556, -0.2501, -0.3282, -0.4963,
-0.3028, -0.5007, -0.4114, -0.4488, -0.6029, -0.4322, -0.5713,
-0.3254, -0.3470, -0.1954, -0.2299, -0.3837, -0.2728, -0.2728,
-0.3137, -0.4674, -0.2508, -0.3944, -0.5503, -0.5259, -0.7031,
-0.4668, -0.4259, -0.1611, -0.3039, -0.7107, -0.5001, -0.3447,
-0.6504, -0.6410, -0.4217, -0.2464, -0.7371, -0.3611, -0.3044,
-0.4561, -0.3025, -0.5268, 0.0381, -0.3804, -0.2355, -0.4848,
-0.5232, -0.5579, 0.1405, -0.3290, -0.0300, -0.3975, -0.5651,
-0.6571, -0.2703, -0.2884, -0.1925, -0.4234, -0.1874, 0.0033,
-0.2659, -0.1195, -0.1413, -0.4644, 0.1231, -0.4137, -0.3051,
-0.4111, -0.3668, -0.1711, -0.5422, -0.5880, -0.4089, -0.4194,
-0.5928, -0.4504, -0.5832, -0.2813, -0.8297, -0.3891, -0.3753,
-0.1404, -0.4839, -0.4172, -0.2163, -0.3893, -0.5516, -0.5795,
-0.4672, -0.4276, -0.3719, -0.5412, -0.1255, -0.2490, -0.5926,
-0.3345, -0.3791, -0.4391, -0.3159, -0.3038, -0.2926, -0.2882,
-0.5251, -0.1011, -0.2111, -0.4174, -0.5177, -0.5118, -0.3595,
-0.4956, -0.4753, -0.5925, -0.6087, -0.4069, -0.4549, -0.1989,
-0.1740, -0.3206, -0.4496, -0.2257, -0.6268, -0.5194, -0.4852,
-0.1086, -0.6572, -0.7426, -0.3687, -0.6185, -0.4834, -0.2706,
-0.5423, -0.4072, -0.0834, -0.2501, -0.4635, -0.4078, -0.5541,
-0.0337, -0.2734, -0.5602, -0.5042, -0.1952, -0.3742, -0.5523,
-0.5093, -0.2505, -0.5155, -0.3092, -0.4025, -0.4081, -0.4622,
-0.4024, -0.6621, -0.2153, -0.6288, -0.2819, -0.1222, -0.2232,
-0.5776, -0.2034, -0.3000, -0.4742, -0.7366, -0.2013, -0.2618,
0.1005, -0.2582, -0.7583, 0.1883, -0.3596, -0.2567, -0.3549,
-0.5619, -0.5463, -0.3409, -0.2691, -0.3938, -0.4213, -0.2406,
-0.5667, -0.1326, -0.3434, -0.6747, -0.2804, -0.2719, -0.5073,
-0.3260, -0.0403, -0.2664, -0.3188, -0.2938, -0.5389, -0.2050,
-0.4609, -0.2776, -0.5544, -0.3783, -0.3962, -0.3812, 0.0367,
-0.4590, -0.4816, -0.3661, -0.2352, -0.0971, -0.6940, -0.3398,
-0.3710, -0.0319, -0.4874, -0.3852, -0.4983, -0.1890, -0.3516,
-0.2117, -0.4780, -0.4530, -0.4071, -0.1249, -0.3350, -0.1764,
-0.3650, -0.7885, -0.3486, -0.3382, -0.5207, -0.0697, -0.3521,
-0.1826, -0.4978, -0.3711, -0.2221, -0.3227, -0.3846, -0.2063,
-0.4818, -0.3935, -0.3178, -0.3863, -0.2697, -0.3244, -0.7194,
-0.3130, -0.4984, -0.3723, -0.3482, -0.4857, -0.0743, -0.4222,
-0.5206, -0.6223, -0.2797, 0.7561, -0.5744, -0.2969, -0.3717,
-0.4671, -0.4312, -0.5382, 0.0194, -0.5450, -0.4912, -0.1807,
-0.2839, -0.4784, -0.6239, -0.4280, -0.3659, -0.0494, -0.7320,
-0.6327, -0.4970, -0.0153, -0.5496, -0.3423, -0.2341, -0.1791,
-0.3122, 0.4985, -0.4146, -0.7272, -0.6907, -0.3604, -0.3958,
-0.0578, -0.5555, 0.1433, -0.3553, -0.3223, -0.2659, -0.3620,
-0.4346, -0.5469, -0.5171, -0.7143, -0.3552, -0.6430, -0.2717,
-0.0772, -0.5156, -0.4646, -0.7085, -0.1570, -0.4476, -0.5665,
-0.3551, -0.4731, -0.6038, -0.4503, -0.3136, -0.6242, -0.4143,
-0.3649, -0.2212, -0.4198, 0.4058, -0.1536, -0.4433, -0.4800,
-0.2788, -0.5187, -0.3629, -0.3651, -0.6057, -0.6628, -0.0567,
-0.2968, -0.1914, -0.2634, -0.2080, -0.6012, -0.3833, -0.7599,
-0.2679, -0.6340, -0.2526, -0.4454, -0.2467, -0.4620, -0.3775,
-0.5725, -0.3658, -0.6906, -0.2763, -0.2170, -0.5583, 2.3687,
-0.5421, -0.2879, -0.4934, -0.3546, -0.3680, -0.4790, 0.0820,
-0.2224, -0.1593, -0.2072, -0.5971, -0.3634, -0.3488, -0.4481,
-0.5325, -0.3482, -0.7042, -0.3770, -0.5317, -0.4193, -0.3078,
-0.4643, -0.0888, -0.1248, -0.2698, -0.4371, -0.4275, -0.4113,
-0.3019, -0.3439, -0.4192, -0.4212, -0.5230, -0.6090, -0.2991,
-0.4603, -0.3159, -0.5463, -0.4443, -0.4015, -0.3053, -0.4116,
-0.5135, -0.1921, -0.5359, 0.1316, -0.4799, -0.6427, -0.1373,
-0.2284, -0.2840, -0.2663, -0.2728, -0.3347, -0.2260, -0.1522,
-0.1046, -0.7010, -0.5304, -0.1956, -0.3509, -0.1656, -0.4885,
-0.2164, -0.5814, -0.4442, -0.1139, -0.5248, -0.2839, -0.3137,
-0.1550, -0.3380, -0.6125, -0.2917, -0.4101, -0.2943, -0.3565,
-0.6214, -0.2557, -0.4759, -0.2127, -0.2462, -0.7725, -0.4395,
-0.2886, -0.3807, -0.2373, -0.2999, -0.2010, -0.4811, -0.0841,
-0.6150, -0.2098, -0.4323, -0.0174, -0.7442, -0.7739, -0.2853,
-0.0977, -0.2996, -0.2086, -0.6203, -0.2199, -0.4272, -0.3907,
-0.4279, -0.5308, -0.2273, -0.3776, -0.3552, -0.3874, -0.4162,
-0.4903, -0.6617, -0.5385, -0.2847, -1.1933, -0.1081, -0.3179,
-0.3225, -0.3558, -0.4197, -0.4201, -0.3552, 0.0195, -0.6408,
-0.4163, -0.4390, -0.0696, -0.2452, -0.4574, 0.0262, -0.2046,
-0.7068, -0.4184, -0.3031, -0.2492, -0.4644, -0.3548, -0.0914,
-0.6132], device='cuda:0')), ('layer4.2.bn1.running_var', tensor([ 0.2818, 0.4219, 0.3018, 0.2590, 0.3269, 0.2878, 0.2967,
0.2464, 0.2859, 0.3018, 0.2984, 0.2983, 0.2660, 0.2536,
0.2764, 0.3949, 0.5205, 0.2936, 0.2643, 0.2821, 0.3480,
0.2967, 0.2740, 0.2172, 0.3074, 0.3526, 0.2807, 0.2198,
0.3585, 0.3047, 0.2974, 0.2654, 0.3064, 0.2697, 0.3678,
0.2415, 0.2867, 0.3166, 0.3278, 0.3558, 0.2935, 0.3585,
0.2349, 0.2485, 0.3449, 0.2864, 0.2880, 0.3269, 0.3399,
0.2436, 0.3230, 0.3042, 0.2659, 0.3521, 0.2868, 0.3270,
0.3389, 0.2897, 0.2815, 0.3419, 0.3103, 0.2668, 0.3615,
0.4319, 0.2364, 0.3301, 0.2826, 0.3409, 0.2542, 0.2639,
0.2732, 0.2965, 0.2664, 0.2569, 0.3908, 0.2379, 0.2801,
0.3468, 0.2559, 0.2810, 0.3900, 0.2971, 0.3096, 0.3226,
0.2793, 0.2519, 0.2662, 0.2736, 0.2143, 0.2563, 0.3099,
0.2380, 0.2758, 0.3284, 0.2718, 0.2912, 0.2425, 0.2711,
0.2935, 0.3186, 0.2499, 0.2681, 0.3769, 0.3171, 0.3602,
0.3800, 0.2601, 0.3153, 0.2917, 0.2392, 0.3571, 0.3507,
0.3352, 0.2890, 0.3227, 0.3786, 0.3357, 0.3133, 0.3320,
0.2699, 0.2761, 0.3223, 0.2849, 0.3304, 0.3434, 0.2293,
0.2695, 0.3275, 0.2656, 0.2385, 0.3365, 0.3999, 0.2643,
0.2783, 0.3565, 0.3221, 0.2396, 0.2628, 0.3279, 0.3879,
0.2876, 0.2333, 0.2444, 0.3668, 0.3386, 0.2584, 0.2588,
0.3093, 0.3518, 0.2918, 0.2980, 0.2759, 0.2561, 0.3206,
0.3230, 0.4120, 0.2672, 0.2964, 0.2999, 0.3328, 0.3152,
0.3423, 0.2524, 0.2435, 0.2971, 0.2814, 0.3407, 0.3218,
0.2156, 0.2685, 0.3287, 0.2934, 0.2981, 0.2786, 0.2879,
0.3161, 0.3858, 0.2636, 0.3891, 0.2517, 0.3026, 0.3331,
0.2540, 0.2728, 0.3279, 0.3254, 0.3256, 0.3264, 0.2598,
0.4030, 0.2742, 0.2412, 0.3805, 0.2611, 0.2193, 0.2187,
0.2932, 0.3674, 0.3098, 0.3760, 0.2729, 0.2807, 0.2590,
0.2401, 0.2415, 0.3242, 0.3648, 0.2899, 0.3350, 0.2201,
0.2476, 0.3251, 0.3098, 0.2942, 0.2960, 0.3151, 0.2203,
0.2527, 0.2555, 0.4135, 0.2723, 0.3708, 0.3663, 0.2759,
0.3996, 0.2789, 0.2726, 0.2192, 0.3107, 0.2921, 0.3371,
0.2909, 0.3394, 0.3460, 0.2890, 0.1997, 0.2589, 0.2899,
0.3066, 0.2376, 0.2531, 0.3339, 0.2450, 0.2919, 0.2979,
0.2964, 0.2607, 0.2144, 0.3884, 0.3285, 0.3440, 0.3098,
0.2187, 0.2985, 0.2415, 0.2437, 0.3215, 0.2542, 0.2866,
0.3445, 0.3550, 0.3831, 0.2532, 0.4206, 0.2788, 0.2873,
0.2581, 0.3022, 0.3245, 0.2983, 0.2905, 0.3249, 0.3033,
0.3408, 0.2754, 0.2455, 0.4535, 0.3811, 0.2467, 0.2661,
0.2567, 0.2944, 0.3651, 0.2844, 0.4748, 0.2388, 0.2845,
0.3010, 0.2868, 0.3006, 0.2273, 0.2714, 0.3327, 0.3486,
0.2524, 0.3999, 0.3172, 0.3575, 0.2758, 0.2879, 0.2363,
0.2601, 0.4610, 0.2992, 0.4502, 0.3743, 0.2784, 0.2094,
0.2980, 0.3214, 0.3571, 0.2364, 0.3652, 0.2606, 0.3328,
0.2867, 0.4058, 0.2934, 0.3427, 0.3060, 0.2979, 0.2768,
0.3077, 0.2453, 0.3531, 0.3516, 0.2245, 0.3254, 0.2251,
0.2922, 0.2305, 0.3327, 0.2904, 0.2473, 0.2787, 0.2665,
0.2986, 0.2678, 0.2955, 0.5841, 0.2484, 0.2963, 0.2876,
0.3828, 0.2584, 0.2698, 0.2748, 0.2830, 0.2602, 0.3291,
0.3441, 0.2681, 0.2507, 0.3112, 0.3943, 0.2855, 0.3273,
0.2784, 0.3442, 0.2305, 0.2324, 0.2816, 0.3615, 0.3077,
0.2752, 0.2907, 0.2878, 0.2704, 0.2301, 0.2767, 1.5347,
0.2679, 0.2760, 0.3591, 0.2608, 0.2477, 0.2395, 0.3706,
0.3808, 0.3060, 0.2512, 0.3388, 0.2292, 0.2867, 0.2333,
0.4648, 0.2752, 0.3366, 0.2545, 0.3243, 0.2871, 0.2620,
0.2558, 0.3114, 0.2634, 0.3283, 0.2643, 0.2793, 0.3617,
0.2522, 0.2685, 0.3646, 0.2652, 0.2206, 0.3004, 0.2228,
0.3569, 0.2911, 0.2652, 0.2559, 0.3240, 0.1925, 0.3038,
0.2987, 0.2300, 0.3607, 0.2756, 0.2875, 0.4537, 0.2203,
0.3042, 0.2334, 0.2884, 0.2506, 0.2656, 0.2788, 0.2651,
0.2877, 0.3483, 0.2809, 0.2668, 0.2940, 0.2399, 0.2325,
0.3089, 0.3369, 0.2797, 0.3085, 0.2398, 0.3313, 0.2474,
0.2123, 0.3346, 0.3084, 0.2624, 0.3253, 0.3226, 0.2541,
0.3615, 0.2382, 0.2480, 0.2786, 0.4920, 0.3639, 0.2836,
0.4313, 0.2418, 0.2604, 0.2590, 0.3016, 0.3283, 0.3391,
0.3306, 0.3608, 0.2795, 0.2840, 0.3057, 0.5184, 0.2214,
0.2838, 0.2713, 0.3211, 0.3641, 0.2636, 0.2851, 0.2947,
0.3401, 0.3733, 0.2256, 0.2555, 0.3183, 0.3163, 0.3159,
0.2294, 0.3082, 0.2465, 0.3001, 1.0733, 0.3644, 0.2629,
0.3280, 0.2766, 0.2734, 0.2561, 0.3235, 0.7038, 0.3213,
0.2612, 0.3848, 0.2959, 0.2666, 0.2350, 0.2649, 0.3245,
0.3183, 0.2460, 0.2821, 0.4127, 0.3772, 0.2798, 0.2922,
0.2971], device='cuda:0')), ('layer4.2.conv2.weight', tensor([[[[ 4.3621e-03, 6.8473e-03, 8.0288e-03],
[ 6.7248e-03, 8.9230e-03, 1.0765e-02],
[ 1.0765e-02, 7.9923e-03, 1.1002e-02]],
[[ 1.0535e-02, 1.0439e-02, 1.3608e-02],
[ 9.3876e-03, 1.4511e-02, 7.1621e-03],
[ 8.3283e-03, 1.1501e-02, 5.5217e-03]],
[[ 5.9216e-03, 4.4101e-03, 5.2177e-03],
[-8.6167e-04, -3.2242e-03, -6.7179e-04],
[ 4.5840e-05, -4.9668e-04, 1.4025e-03]],
...,
[[-4.1221e-03, 7.0039e-03, 3.8310e-03],
[ 3.8982e-03, 1.1210e-02, 1.0156e-02],
[ 5.6188e-03, 1.7809e-02, 1.1437e-02]],
[[-9.4583e-04, -1.2922e-03, -3.5472e-03],
[ 1.2131e-03, 2.9635e-04, -3.1784e-03],
[-7.3291e-03, -8.8249e-03, -1.1508e-02]],
[[-6.6200e-03, -9.3631e-03, -6.3588e-03],
[-5.2284e-04, 6.7514e-04, 4.8125e-03],
[-1.5984e-03, -1.2874e-03, 1.0811e-03]]],
[[[-7.3626e-03, -3.9822e-03, -7.4166e-03],
[-5.5884e-03, -1.4974e-03, -4.5140e-03],
[-4.6087e-03, 1.4543e-03, -4.4360e-03]],
[[ 9.1032e-03, 1.0345e-02, 1.3476e-02],
[ 3.4463e-03, 7.2927e-03, 9.1789e-03],
[ 2.4254e-03, 7.0794e-03, 6.4320e-03]],
[[ 1.2688e-02, 1.9715e-02, 1.8962e-02],
[ 1.3353e-02, 1.5823e-02, 1.7648e-02],
[ 1.3564e-02, 1.6767e-02, 2.0256e-02]],
...,
[[-6.6037e-03, -9.7982e-03, -1.0898e-02],
[-6.0535e-03, -6.4405e-03, -5.4698e-03],
[ 5.0908e-03, 3.4297e-03, 5.3977e-03]],
[[-5.5245e-03, -1.0339e-02, -5.4738e-03],
[-7.5820e-03, -7.6047e-03, -5.4363e-03],
[-4.6174e-03, -5.5401e-03, -4.5972e-03]],
[[ 6.9310e-04, 1.0304e-03, -2.0162e-03],
[ 2.9895e-03, 7.5537e-03, 3.2833e-03],
[-2.2063e-03, 1.8036e-03, -6.4289e-03]]],
[[[-3.4324e-03, -5.4486e-03, -9.6182e-03],
[-1.0760e-02, -9.2769e-03, -1.2156e-02],
[-8.7225e-03, -8.7186e-03, -9.4842e-03]],
[[ 2.7660e-02, 2.2491e-02, 3.1649e-02],
[ 2.0971e-02, 1.5626e-02, 2.4904e-02],
[ 1.7404e-02, 1.5938e-02, 2.1827e-02]],
[[ 1.5516e-02, 1.2521e-02, 1.7123e-02],
[ 1.6623e-02, 5.7430e-03, 1.5497e-02],
[ 1.7264e-02, 1.0979e-02, 1.4872e-02]],
...,
[[-1.0173e-02, -1.2509e-02, -1.2630e-02],
[-6.0805e-03, -8.9648e-03, -1.0305e-02],
[-1.0036e-02, -1.3899e-02, -1.2060e-02]],
[[-1.0807e-02, -1.1518e-02, -8.1304e-03],
[-8.4999e-03, -9.2836e-03, -4.6052e-03],
[-5.2926e-03, -1.2375e-02, -8.8108e-03]],
[[-1.8685e-02, -2.2845e-02, -1.9386e-02],
[-1.3066e-02, -1.1810e-02, -1.3100e-02],
[-1.1887e-02, -1.3642e-02, -1.0794e-02]]],
...,
[[[ 3.1914e-03, 3.8416e-03, 4.5126e-04],
[ 8.4163e-04, 1.0403e-03, 2.0931e-03],
[-3.4614e-03, -3.5079e-03, -7.2330e-04]],
[[-1.0527e-02, -1.2018e-02, -1.3126e-02],
[-6.9252e-03, -4.5654e-03, -8.1068e-03],
[-2.5715e-04, 3.0966e-03, -4.3668e-03]],
[[-3.0327e-04, 7.0920e-04, 6.0681e-04],
[ 4.3166e-03, 2.8169e-03, 2.7342e-03],
[ 5.8915e-03, 1.6588e-03, 2.5264e-03]],
...,
[[ 2.2021e-03, 5.6848e-03, 6.0258e-03],
[ 1.1379e-02, 1.8494e-02, 2.0603e-02],
[ 5.7015e-03, 1.2468e-02, 1.4021e-02]],
[[ 4.4007e-03, 2.5775e-03, 6.7226e-03],
[-1.4912e-03, -2.2758e-03, 1.6108e-03],
[-6.9917e-03, -4.0513e-03, -4.4224e-03]],
[[ 2.7756e-02, 2.2483e-02, 3.3769e-02],
[ 2.4960e-02, 1.7393e-02, 3.0202e-02],
[ 2.8441e-02, 1.9855e-02, 3.0437e-02]]],
[[[-5.7504e-03, -1.7024e-02, -1.6196e-02],
[-9.5403e-03, -1.3252e-02, -1.7009e-02],
[-1.1870e-02, -1.2318e-02, -1.3297e-02]],
[[ 2.0091e-02, 1.7347e-02, 1.5497e-02],
[ 2.1621e-02, 1.3661e-02, 1.9919e-02],
[ 2.0871e-02, 1.8383e-02, 2.0216e-02]],
[[-5.7679e-03, -2.8263e-03, -4.3957e-03],
[-2.2350e-03, -1.9012e-03, -4.7539e-03],
[-6.1207e-03, -4.4888e-03, -5.5439e-03]],
...,
[[ 1.0894e-02, 8.9690e-03, 6.2257e-03],
[ 1.0858e-02, 8.5993e-03, 7.6358e-03],
[ 3.1347e-03, 3.8935e-03, -2.2421e-03]],
[[-1.0813e-02, -1.2291e-02, -9.8513e-03],
[-2.0932e-03, -4.8284e-03, -7.5115e-04],
[-1.7459e-03, -3.4481e-03, -6.6017e-04]],
[[-5.9274e-03, -5.1394e-03, -8.1561e-03],
[-5.1022e-03, -5.8396e-03, -3.8814e-03],
[-3.1284e-03, -3.8703e-03, -1.7811e-03]]],
[[[ 1.6075e-03, 4.6062e-04, -2.0657e-03],
[-4.1777e-03, -3.7883e-03, -2.4825e-03],
[-2.9089e-03, -2.9392e-03, -3.3209e-03]],
[[ 1.1568e-02, 1.1682e-02, 1.0651e-02],
[ 1.2498e-02, 8.0237e-03, 1.4024e-02],
[ 1.6738e-02, 1.7037e-02, 1.7864e-02]],
[[-2.8646e-03, -1.2487e-02, -1.1430e-02],
[-4.1906e-03, -4.6766e-03, -1.1029e-02],
[-6.0544e-03, -8.4252e-03, -1.3997e-02]],
...,
[[ 1.0390e-03, 2.1785e-03, -3.9463e-03],
[-2.8940e-04, -3.1756e-03, -8.1843e-03],
[-5.2049e-03, -6.1339e-03, -4.9104e-03]],
[[-1.6184e-03, -3.3409e-03, -1.7560e-03],
[-6.3296e-03, -3.5010e-03, 3.2006e-04],
[-7.2348e-04, 3.3084e-03, -6.8654e-04]],
[[-1.9266e-03, -5.5553e-03, -6.5790e-03],
[-6.1098e-03, -1.1194e-02, -1.2170e-02],
[-1.3577e-02, -1.4458e-02, -1.8718e-02]]]], device='cuda:0')), ('layer4.2.bn2.weight', tensor([ 0.1960, 0.1666, 0.1982, 0.2569, 0.2707, 0.1756, 0.2237,
0.2237, 0.2370, 0.2180, 0.1902, 0.2278, 0.2278, 0.1883,
0.2098, 0.2905, 0.2305, 0.1993, 0.2083, 0.2108, 0.2312,
0.2131, 0.1651, 0.2562, 0.2190, 0.2358, 0.2198, 0.2041,
0.2156, 0.2367, 0.2431, 0.1376, 0.2873, 0.2147, 0.3033,
0.1875, 0.2236, 0.2337, 0.2504, 0.2051, 0.1765, 0.2041,
0.2226, 0.1875, 0.2221, 0.2119, 0.2020, 0.2119, 0.2634,
0.2388, 0.2098, 0.1667, 0.2704, 0.1934, 0.2579, 0.1799,
0.2296, 0.2026, 0.2150, 0.2267, 0.2251, 0.2376, 0.1989,
0.2332, 0.2316, 0.1681, 0.1983, 0.1239, 0.2375, 0.2932,
0.2494, 0.1826, 0.2251, 0.2453, 0.2338, 0.2176, 0.3047,
0.2032, 0.1896, 0.2153, 0.1966, 0.2276, 0.1678, 0.1983,
0.2521, 0.1930, 0.1375, 0.1905, 0.1750, 0.2550, 0.1783,
0.2282, 0.2110, 0.2518, 0.2366, 0.2550, 0.2185, 0.2157,
0.2375, 0.2473, 0.1644, 0.2347, 0.2374, 0.1854, 0.2326,
0.2472, 0.1819, 0.2360, 0.2137, 0.1847, 0.2615, 0.1840,
0.2663, 0.1607, 0.1960, 0.2315, 0.1842, 0.2383, 0.1705,
0.2243, 0.1638, 0.2129, 0.2198, 0.1938, 0.1995, 0.2418,
0.2547, 0.2060, 0.2036, 0.2617, 0.1417, 0.2040, 0.2211,
0.1723, 0.1955, 0.2540, 0.2521, 0.2221, 0.2112, 0.2042,
0.2014, 0.2533, 0.1656, 0.1502, 0.2273, 0.2342, 0.2556,
0.2271, 0.1900, 0.2050, 0.2277, 0.2250, 0.2557, 0.2381,
0.1804, 0.1982, 0.2209, 0.2331, 0.2109, 0.2913, 0.1498,
0.2041, 0.2156, 0.1679, 0.2597, 0.2253, 0.2693, 0.2387,
0.1961, 0.2420, 0.2384, 0.2458, 0.2211, 0.2932, 0.2951,
0.1981, 0.2416, 0.2284, 0.2648, 0.2235, 0.2046, 0.2380,
0.2477, 0.2331, 0.2053, 0.2558, 0.2410, 0.2291, 0.2740,
0.2283, 0.1862, 0.1735, 0.2579, 0.2294, 0.1748, 0.1785,
0.2053, 0.2097, 0.2328, 0.1889, 0.1705, 0.1874, 0.2009,
0.1794, 0.2308, 0.1647, 0.1702, 0.1293, 0.1670, 0.1653,
0.1912, 0.1990, 0.2008, 0.2471, 0.2336, 0.2955, 0.2585,
0.2039, 0.2182, 0.2140, 0.1776, 0.1662, 0.1507, 0.2171,
0.1887, 0.2305, 0.1936, 0.2743, 0.2642, 0.1527, 0.2048,
0.1576, 0.1427, 0.1859, 0.2236, 0.2216, 0.2304, 0.1676,
0.1575, 0.2388, 0.1454, 0.2081, 0.2506, 0.2043, 0.1739,
0.2476, 0.2416, 0.2223, 0.1906, 0.1802, 0.2520, 0.2215,
0.2122, 0.1803, 0.2093, 0.2201, 0.2575, 0.2249, 0.2041,
0.2198, 0.1858, 0.1532, 0.1891, 0.2496, 0.1669, 0.1939,
0.2283, 0.2247, 0.2430, 0.2464, 0.2374, 0.1529, 0.1666,
0.1893, 0.1902, 0.2099, 0.1445, 0.1866, 0.2019, 0.2318,
0.2388, 0.2113, 0.1621, 0.2279, 0.2715, 0.1986, 0.2229,
0.2274, 0.2348, 0.2481, 0.2082, 0.1804, 0.2191, 0.2911,
0.2412, 0.2109, 0.1949, 0.2112, 0.1736, 0.2333, 0.1943,
0.1677, 0.2038, 0.2099, 0.2121, 0.2196, 0.1956, 0.2355,
0.1762, 0.2179, 0.2003, 0.1680, 0.2147, 0.1937, 0.2241,
0.2143, 0.2318, 0.2209, 0.1857, 0.2263, 0.2404, 0.1897,
0.1783, 0.2169, 0.2187, 0.2242, 0.2493, 0.2395, 0.2327,
0.2695, 0.2592, 0.2772, 0.2413, 0.2287, 0.1621, 0.2077,
0.2362, 0.2276, 0.2203, 0.2527, 0.1873, 0.2077, 0.1972,
0.2227, 0.1943, 0.1677, 0.2390, 0.2414, 0.2757, 0.1878,
0.2126, 0.2082, 0.2301, 0.2165, 0.2041, 0.2465, 0.1819,
0.2222, 0.1736, 0.2763, 0.1816, 0.2324, 0.2088, 0.2357,
0.2147, 0.1938, 0.2403, 0.2039, 0.2348, 0.1830, 0.2098,
0.2449, 0.2228, 0.2592, 0.1693, 0.2282, 0.2288, 0.2398,
0.2518, 0.2329, 0.2540, 0.2379, 0.2481, 0.2066, 0.2446,
0.2810, 0.2626, 0.1923, 0.2103, 0.1324, 0.1498, 0.2304,
0.2391, 0.2396, 0.2144, 0.2297, 0.1993, 0.1411, 0.2558,
0.1949, 0.2400, 0.2178, 0.2171, 0.2420, 0.2537, 0.1849,
0.2402, 0.2269, 0.1609, 0.2113, 0.2467, 0.2895, 0.1772,
0.2170, 0.2771, 0.1837, 0.2797, 0.2177, 0.1730, 0.2241,
0.2053, 0.1916, 0.1934, 0.2121, 0.1933, 0.2088, 0.2343,
0.2156, 0.2124, 0.2108, 0.1756, 0.2634, 0.2473, 0.2481,
0.2933, 0.2481, 0.1821, 0.2054, 0.1696, 0.1910, 0.2360,
0.2633, 0.1921, 0.1925, 0.2063, 0.1860, 0.1856, 0.2095,
0.2353, 0.2580, 0.2080, 0.1798, 0.2350, 0.2684, 0.1926,
0.2113, 0.2075, 0.2645, 0.1766, 0.2122, 0.1696, 0.2254,
0.2106, 0.2232, 0.2472, 0.1907, 0.2479, 0.2324, 0.1957,
0.2047, 0.1911, 0.2319, 0.2098, 0.2127, 0.2661, 0.2514,
0.1636, 0.2059, 0.2236, 0.2489, 0.1550, 0.1825, 0.1739,
0.1996, 0.2281, 0.1917, 0.2028, 0.2126, 0.2268, 0.2324,
0.2144, 0.2430, 0.1333, 0.1699, 0.2379, 0.1829, 0.3097,
0.2505, 0.2276, 0.2269, 0.2311, 0.2134, 0.1934, 0.2087,
0.2258, 0.2343, 0.2014, 0.2107, 0.1827, 0.2268, 0.1902,
0.1932], device='cuda:0')), ('layer4.2.bn2.bias', tensor([-0.1243, -0.0183, -0.0429, -0.1141, -0.1424, -0.0362, -0.1703,
-0.1503, -0.1189, -0.1253, -0.0238, -0.1806, -0.1130, -0.0715,
-0.0503, -0.1961, -0.0958, -0.0567, -0.1074, -0.0129, 0.0017,
-0.0265, -0.0474, -0.1157, -0.1570, -0.1506, -0.0678, -0.0520,
-0.0641, -0.0893, -0.0423, 0.0188, -0.2314, -0.1541, -0.1288,
-0.0750, -0.0349, -0.0592, -0.1491, -0.0431, -0.0539, -0.0696,
-0.0910, -0.1388, -0.1124, -0.1067, -0.0555, -0.1099, -0.0880,
-0.0500, -0.0526, -0.0424, -0.0801, -0.1218, -0.0228, -0.0871,
-0.1109, -0.0983, -0.0626, -0.1034, -0.0962, -0.0747, -0.0448,
-0.0471, -0.0606, -0.0058, -0.0420, 0.2089, -0.0857, -0.1062,
-0.0945, -0.0307, -0.1210, -0.0471, -0.0478, -0.1035, -0.1271,
-0.0637, -0.1102, -0.0691, -0.0938, -0.0167, -0.0611, -0.1283,
-0.1327, -0.0763, 0.0004, -0.0930, -0.0773, -0.1016, -0.0877,
-0.1463, -0.1099, -0.0632, -0.1236, -0.0734, -0.0836, -0.1389,
-0.1660, -0.1402, -0.0272, -0.1296, -0.0836, -0.0497, -0.0669,
-0.1116, -0.0565, -0.0552, -0.1291, -0.0933, -0.1058, -0.0260,
-0.0931, -0.0716, -0.0935, -0.0441, -0.0083, -0.1687, -0.0279,
-0.1149, -0.0220, -0.0753, -0.0942, -0.0551, -0.1198, -0.1593,
-0.1117, -0.1108, -0.0710, -0.0416, -0.0347, -0.0572, -0.0870,
-0.0545, -0.0461, -0.0493, -0.0684, -0.0505, -0.0353, -0.1249,
0.0247, -0.1704, -0.0394, -0.0424, -0.0794, -0.0874, -0.1318,
-0.0462, -0.1119, -0.0684, -0.1006, -0.1061, -0.0418, -0.0965,
-0.0129, -0.0330, -0.0302, -0.0930, -0.0976, -0.2124, -0.0811,
-0.1556, -0.1047, -0.0993, -0.1133, -0.0447, -0.0891, -0.0199,
-0.0624, -0.0444, -0.1126, -0.1160, -0.1264, -0.2631, -0.1851,
-0.0902, -0.1165, -0.1596, -0.0862, -0.0878, -0.0385, -0.0840,
-0.1490, -0.1032, -0.0966, -0.1597, -0.0508, -0.1611, -0.1136,
-0.1472, -0.0534, -0.0823, -0.0625, -0.1578, -0.0523, -0.0670,
-0.1088, -0.0667, -0.0216, -0.0758, -0.0114, -0.0948, -0.0463,
-0.0516, -0.0314, -0.0631, -0.0828, -0.0450, 0.0133, -0.0596,
-0.1108, 0.0429, -0.0836, -0.1203, -0.0839, -0.1680, -0.1997,
-0.0196, -0.1713, -0.1337, -0.0643, -0.0289, -0.0490, -0.1013,
-0.0442, -0.0450, -0.0406, -0.1507, -0.1196, -0.0131, -0.0744,
-0.0731, -0.0333, -0.0145, -0.0739, -0.0895, -0.0823, -0.0303,
-0.0353, -0.0151, -0.0295, -0.0327, -0.0917, -0.0815, 0.0971,
-0.1531, -0.0581, -0.1546, -0.1214, -0.0791, -0.1144, -0.0997,
-0.0971, -0.0676, -0.0924, -0.0585, -0.2099, -0.1759, -0.1159,
-0.0895, -0.0744, -0.0253, -0.0829, -0.1205, -0.0170, -0.0427,
-0.0418, -0.0712, -0.0768, -0.1305, -0.2116, -0.0212, -0.0443,
-0.0673, -0.0817, -0.1166, 0.0013, -0.0883, -0.0945, -0.1882,
-0.1002, -0.1185, -0.0526, -0.0354, -0.0257, -0.0577, -0.0585,
-0.0793, -0.1257, -0.0266, -0.0594, 0.0277, -0.0003, -0.1008,
-0.0561, -0.0576, -0.0975, -0.0811, -0.0339, -0.0195, -0.0402,
-0.0712, -0.0397, -0.1024, -0.0680, -0.0604, -0.0823, -0.1081,
-0.0618, -0.1450, -0.0662, -0.0301, -0.1006, -0.0605, -0.0694,
-0.0779, -0.1030, -0.1095, -0.0663, -0.0704, -0.0575, -0.0875,
0.0154, -0.0593, -0.0919, -0.0702, -0.0673, -0.0299, -0.1807,
-0.1366, -0.0730, -0.0598, -0.1205, -0.1731, -0.0354, -0.0688,
-0.0522, -0.0998, -0.0821, -0.1027, -0.0612, -0.0953, -0.0503,
-0.1068, -0.1060, -0.0619, -0.0776, -0.0385, -0.0757, -0.0015,
-0.1243, -0.1346, -0.0363, -0.1065, -0.0482, -0.1114, -0.0098,
-0.0781, -0.0283, -0.1947, -0.0791, -0.0508, -0.0074, -0.0986,
-0.1598, -0.0054, -0.0967, -0.1063, -0.1627, -0.0264, -0.0532,
-0.0851, -0.0261, -0.1193, -0.0394, -0.0787, -0.0842, -0.0644,
-0.0397, -0.0730, -0.1352, -0.1144, -0.1592, -0.0565, -0.1612,
-0.1364, -0.0599, -0.0337, -0.1010, -0.0368, 0.0109, -0.0652,
-0.0716, -0.1679, -0.1586, -0.0247, -0.0882, -0.0202, -0.1415,
-0.0617, -0.1208, -0.0598, -0.0936, -0.1198, -0.0384, -0.0792,
-0.0958, -0.1251, -0.0387, -0.0432, -0.0643, -0.0651, -0.0735,
-0.0445, -0.0781, -0.0249, -0.0766, -0.1394, -0.0719, -0.1423,
-0.0669, -0.0506, -0.0714, -0.1012, -0.1009, -0.0537, -0.1678,
-0.0870, -0.0411, -0.1189, -0.0587, -0.2393, -0.0928, -0.0790,
-0.1138, -0.1480, -0.0886, -0.0478, -0.0455, -0.0398, -0.0539,
-0.0697, -0.0649, -0.0922, -0.0348, -0.0729, -0.1207, -0.0838,
-0.1191, -0.0977, -0.0858, -0.0330, -0.1276, -0.1383, -0.0470,
-0.1554, -0.0255, -0.1045, -0.0444, -0.0745, -0.0645, 0.0115,
-0.0215, -0.0489, -0.1513, -0.0512, -0.1063, -0.1026, -0.0181,
-0.1060, -0.0874, -0.0564, -0.0690, -0.1142, -0.2051, -0.0734,
-0.0202, 0.0134, -0.0781, -0.1180, -0.0505, -0.0418, -0.0414,
-0.0433, -0.0851, -0.0386, -0.1022, -0.1161, -0.0725, -0.0858,
-0.1376, -0.0938, -0.0120, -0.0202, -0.1104, -0.0841, -0.3079,
-0.0816, -0.0703, -0.0657, -0.1166, -0.1218, -0.0233, -0.1109,
-0.1061, -0.1438, -0.0817, -0.0771, -0.0646, -0.0804, -0.0445,
-0.1100], device='cuda:0')), ('layer4.2.bn2.running_mean', tensor([-0.1406, -0.1151, -0.0995, -0.0862, -0.0908, -0.0808, -0.1409,
-0.1233, -0.0633, -0.1089, -0.2006, -0.1683, -0.0729, -0.1228,
-0.0853, -0.1277, -0.1239, -0.0986, -0.1392, -0.0763, -0.0770,
-0.1021, -0.0835, -0.0940, -0.1496, -0.1001, -0.0825, -0.0864,
-0.0911, -0.0889, -0.0907, -0.1028, -0.1154, -0.0628, -0.0945,
-0.1258, -0.0862, -0.0777, -0.1398, -0.0887, -0.1294, -0.1244,
-0.0981, -0.1366, -0.1043, -0.1549, -0.0617, -0.1268, -0.1082,
-0.0909, -0.0962, -0.1498, -0.1086, -0.1518, -0.0999, -0.1226,
-0.1280, -0.0837, -0.0951, -0.0867, -0.0841, -0.1042, -0.0778,
-0.0810, -0.0974, -0.1322, -0.0909, -0.1023, -0.1257, -0.1030,
-0.0510, -0.0762, -0.0936, -0.0587, -0.0984, -0.1270, -0.0900,
-0.1275, -0.1378, -0.1345, -0.1095, -0.0942, -0.1391, -0.1263,
-0.0878, -0.1048, -0.1147, -0.0991, -0.1335, -0.1033, -0.1458,
-0.1099, -0.0971, -0.0637, -0.1347, -0.0939, -0.0873, -0.0675,
-0.1786, -0.1286, -0.0460, -0.1147, -0.1119, -0.0975, -0.0853,
-0.1363, -0.1601, -0.1504, -0.0977, -0.1138, -0.1005, -0.1342,
-0.1069, -0.1453, -0.0976, -0.0862, -0.1251, -0.1240, -0.1103,
-0.1341, -0.0788, -0.1167, -0.0946, -0.0928, -0.1286, -0.0906,
-0.1330, -0.1156, -0.1153, -0.0859, -0.1068, -0.0869, -0.1196,
-0.1204, -0.0989, -0.0734, -0.0797, -0.1367, -0.0987, -0.1467,
-0.0962, -0.0897, -0.0559, -0.0937, -0.0991, -0.1184, -0.0860,
-0.1000, -0.1491, -0.0787, -0.1063, -0.1121, -0.0918, -0.1272,
-0.0986, -0.0312, -0.0931, -0.1021, -0.1121, -0.1246, -0.1168,
-0.1789, -0.1205, -0.1383, -0.1213, -0.0770, -0.0668, -0.0706,
-0.1163, -0.0559, -0.1153, -0.0833, -0.1010, -0.1402, -0.1138,
-0.1268, -0.0900, -0.1494, -0.1233, -0.1141, -0.0647, -0.1199,
-0.0938, -0.1158, -0.1254, -0.1360, -0.0730, -0.0795, -0.1096,
-0.1631, -0.1322, -0.1874, -0.1232, -0.1258, -0.0693, -0.1183,
-0.0996, -0.1094, -0.0998, -0.1097, -0.1125, -0.0778, -0.0515,
-0.1058, -0.0978, -0.1305, -0.1252, -0.1197, -0.0936, -0.1172,
-0.1055, -0.1395, -0.0785, -0.0927, -0.0697, -0.0964, -0.1509,
-0.0638, -0.1562, -0.1283, -0.1525, -0.1103, -0.1320, -0.1045,
-0.0585, -0.1015, -0.1435, -0.1107, -0.1190, -0.1068, -0.0984,
-0.1379, -0.0934, -0.0799, -0.1075, -0.1136, -0.0727, -0.0986,
-0.1282, -0.0778, -0.1105, -0.0887, -0.1087, -0.1184, -0.0918,
-0.1228, -0.0987, -0.0921, -0.1201, -0.1155, -0.0798, -0.1339,
-0.1225, -0.1113, -0.0970, -0.0833, -0.1606, -0.1364, -0.1232,
-0.0861, -0.1071, -0.0869, -0.0821, -0.1542, -0.0944, -0.0921,
-0.0808, -0.0843, -0.1053, -0.0895, -0.1491, -0.1466, -0.0982,
-0.1009, -0.0933, -0.1211, -0.0916, -0.1238, -0.0883, -0.1406,
-0.0943, -0.1298, -0.0784, -0.0792, -0.0720, -0.0584, -0.0956,
-0.0985, -0.0863, -0.1136, -0.1077, -0.0759, -0.0837, -0.1323,
-0.1098, -0.0965, -0.1303, -0.1080, -0.0923, -0.0624, -0.1383,
-0.1116, -0.0991, -0.0938, -0.0815, -0.1127, -0.1222, -0.1165,
-0.1017, -0.1182, -0.1251, -0.0853, -0.0934, -0.1004, -0.0919,
-0.0953, -0.1560, -0.1123, -0.1326, -0.1109, -0.1250, -0.1541,
-0.0964, -0.1021, -0.0772, -0.0744, -0.0938, -0.0794, -0.1365,
-0.1148, -0.1195, -0.0988, -0.1558, -0.1825, -0.0761, -0.0917,
-0.0891, -0.1105, -0.1009, -0.0981, -0.0836, -0.0991, -0.0813,
-0.1883, -0.0966, -0.0767, -0.1306, -0.0804, -0.1198, -0.0812,
-0.1300, -0.0984, -0.1136, -0.0824, -0.0880, -0.0998, -0.0872,
-0.0908, -0.0936, -0.1472, -0.1095, -0.0938, -0.0818, -0.0751,
-0.1122, -0.0931, -0.1036, -0.1270, -0.1444, -0.1091, -0.1091,
-0.1251, -0.0912, -0.0709, -0.1412, -0.0922, -0.0807, -0.0801,
-0.0662, -0.1239, -0.1783, -0.1215, -0.1209, -0.0773, -0.1378,
-0.0798, -0.1277, -0.0862, -0.0814, -0.1155, -0.1376, -0.1190,
-0.1006, -0.1489, -0.1356, -0.1031, -0.0825, -0.1413, -0.1115,
-0.1149, -0.1143, -0.0744, -0.1010, -0.1161, -0.1490, -0.0958,
-0.0892, -0.1173, -0.1154, -0.0852, -0.0899, -0.0740, -0.1296,
-0.1059, -0.1006, -0.1169, -0.1089, -0.1440, -0.1928, -0.1309,
-0.0826, -0.0912, -0.0846, -0.1184, -0.1321, -0.1144, -0.0986,
-0.1358, -0.0504, -0.1567, -0.1388, -0.1712, -0.1292, -0.0806,
-0.0896, -0.1015, -0.0800, -0.0985, -0.0847, -0.1478, -0.1030,
-0.0844, -0.1325, -0.0903, -0.0722, -0.1086, -0.1557, -0.0813,
-0.1326, -0.0951, -0.1559, -0.0900, -0.0753, -0.1065, -0.0722,
-0.1344, -0.1053, -0.0803, -0.0964, -0.0967, -0.1256, -0.1127,
-0.1133, -0.0938, -0.1039, -0.0600, -0.0820, -0.0876, -0.0867,
-0.0650, -0.1212, -0.0644, -0.1064, -0.1280, -0.1453, -0.1007,
-0.0750, -0.0619, -0.0990, -0.1058, -0.1295, -0.0814, -0.0775,
-0.0990, -0.1252, -0.0561, -0.1267, -0.1314, -0.1097, -0.1178,
-0.1301, -0.0813, -0.1609, -0.0975, -0.0983, -0.1439, -0.1716,
-0.1007, -0.1157, -0.1027, -0.1278, -0.1019, -0.0395, -0.1343,
-0.1023, -0.1201, -0.1075, -0.0920, -0.1078, -0.1261, -0.1325,
-0.0926], device='cuda:0')), ('layer4.2.bn2.running_var', tensor(1.00000e-02 *
[ 2.1623, 2.0630, 1.9764, 1.3487, 1.4731, 1.5967, 1.5314,
1.4749, 1.1172, 1.8753, 3.2347, 2.7457, 1.2883, 1.7666,
1.4367, 2.0526, 1.8535, 1.4150, 2.0861, 1.8450, 1.3630,
2.3285, 1.4668, 1.5944, 2.1772, 1.0556, 1.2544, 1.6050,
1.7178, 1.5833, 2.0783, 1.8257, 1.6892, 0.7774, 1.6076,
1.8570, 1.7204, 1.7327, 2.2477, 1.5197, 1.7597, 1.9239,
1.9414, 1.5278, 1.9690, 2.1297, 1.7173, 2.2682, 1.7278,
1.4870, 1.6254, 2.1016, 1.7982, 2.2601, 1.5973, 2.4091,
1.7760, 1.3839, 1.5534, 1.4231, 1.1256, 1.8655, 1.3469,
1.5241, 1.7810, 2.5238, 1.6843, 3.0373, 2.2320, 1.8842,
1.1811, 1.3621, 1.5652, 1.3270, 2.2138, 2.3542, 1.2544,
2.6215, 1.8797, 2.1231, 1.5540, 2.1180, 2.0486, 1.3492,
1.4505, 1.5379, 2.2064, 1.6448, 1.9576, 1.5124, 2.3914,
1.5889, 1.3677, 1.4119, 1.7964, 2.0140, 1.1570, 1.0748,
2.4479, 1.8752, 1.0515, 1.8256, 2.1243, 1.5431, 1.4566,
2.2047, 1.9633, 1.9586, 1.3455, 1.3346, 1.3855, 2.0746,
1.6787, 2.0011, 1.3967, 1.7237, 1.8949, 1.3135, 1.9393,
1.8389, 1.6295, 1.5267, 1.7266, 1.5523, 1.6695, 1.5095,
2.7887, 1.7227, 1.6187, 1.7977, 1.9080, 1.7052, 1.8784,
1.8933, 1.7498, 1.4290, 1.5954, 1.8162, 1.4685, 1.7935,
1.9826, 1.5455, 1.4736, 1.6087, 1.7292, 1.7192, 1.3326,
2.1388, 1.8093, 1.2517, 1.9401, 2.0663, 2.1021, 2.0142,
2.0140, 0.8883, 1.8043, 1.6623, 1.5213, 1.8791, 1.6236,
2.4187, 1.8110, 1.5273, 1.4968, 1.7515, 1.1819, 1.3653,
1.6529, 1.2268, 1.7231, 1.4763, 1.4203, 1.8191, 2.1736,
1.8911, 1.6863, 2.0536, 2.1010, 2.1134, 1.2853, 1.6106,
1.3885, 1.6530, 1.6543, 1.8800, 1.6990, 0.9903, 2.1764,
1.9859, 1.7589, 2.3106, 2.4774, 1.6287, 1.2255, 1.3664,
1.3211, 1.5527, 1.9333, 2.0541, 2.0767, 1.3803, 1.5019,
1.7731, 1.5820, 2.0479, 1.9356, 1.4933, 1.5922, 1.6329,
1.4340, 2.6186, 1.4020, 1.5658, 1.3237, 1.4407, 2.3472,
1.2982, 1.6605, 1.4921, 2.1637, 1.5369, 1.7565, 1.2678,
1.3431, 2.3217, 2.2253, 1.8673, 1.6687, 2.2166, 1.1738,
1.7736, 1.4437, 1.3277, 1.8628, 2.1177, 1.7201, 1.1197,
2.3232, 2.0424, 1.9846, 1.6314, 1.5005, 1.5962, 2.0749,
2.0260, 1.6364, 1.2829, 1.7929, 1.7173, 1.1236, 1.9974,
1.8551, 2.1022, 1.6858, 1.6529, 1.8231, 1.9987, 1.5378,
1.6795, 1.8671, 2.1971, 1.4412, 2.6014, 2.2483, 1.3138,
1.4205, 1.4397, 1.8046, 1.4214, 2.1349, 2.5215, 1.9378,
1.5399, 1.3274, 2.1644, 1.9600, 2.1405, 1.2731, 1.6397,
1.5325, 2.1190, 1.1257, 1.5243, 1.6768, 1.6839, 1.6199,
1.8855, 1.6706, 1.9141, 1.7416, 1.9491, 1.6838, 2.5401,
1.6852, 1.3966, 2.0999, 1.9428, 1.5853, 1.4320, 2.1688,
1.4990, 1.9478, 1.0340, 1.5176, 1.4692, 1.8177, 1.7415,
1.4373, 2.0562, 1.7297, 1.5283, 1.6115, 2.1027, 1.7120,
1.7027, 1.9805, 1.4616, 2.0333, 1.7171, 2.2584, 2.4591,
1.7715, 1.5800, 1.2999, 1.6333, 2.2549, 1.6388, 1.6892,
2.2079, 1.7974, 1.7147, 2.8181, 2.3451, 1.2792, 1.4806,
1.9881, 1.6279, 1.7011, 1.9351, 1.2076, 1.2986, 1.4025,
2.8958, 1.2385, 1.2384, 2.2270, 1.8794, 1.8077, 1.8602,
2.4110, 1.8218, 2.3037, 1.5904, 1.1965, 1.2727, 1.5887,
1.5687, 1.9292, 2.4336, 1.7648, 1.3068, 1.5439, 1.1572,
1.1976, 1.8226, 2.0702, 1.7879, 2.3361, 1.4940, 1.8620,
1.6266, 1.7844, 1.1632, 1.9173, 1.5001, 1.4802, 1.7057,
1.3982, 2.1183, 2.9495, 1.9882, 1.4935, 1.3258, 1.6732,
1.5070, 2.3543, 1.8544, 1.5772, 1.6006, 2.3366, 2.0796,
1.7667, 2.3090, 1.7315, 1.9190, 1.1993, 2.2927, 1.7147,
1.6660, 1.5692, 1.5964, 1.5265, 2.0541, 2.7905, 1.6777,
1.4124, 1.8153, 1.9187, 1.4134, 1.8619, 2.1103, 1.7711,
1.8866, 1.8439, 2.8168, 1.6748, 1.7051, 2.2196, 1.8671,
1.3445, 1.5431, 1.6157, 1.7293, 2.2116, 2.5098, 1.7752,
2.1298, 1.5942, 2.2351, 1.7555, 2.1551, 2.5215, 1.4470,
1.3808, 1.4425, 1.4821, 2.0759, 1.8270, 2.1669, 1.6184,
1.8180, 1.7534, 1.1066, 1.6903, 1.8573, 2.1007, 1.2092,
1.7944, 1.8708, 2.3229, 1.4168, 1.3010, 1.6433, 1.5890,
1.3439, 1.7035, 1.3820, 1.3819, 1.8980, 1.7616, 2.0269,
2.6833, 1.8221, 1.4641, 0.9924, 1.9356, 1.4457, 1.9390,
1.2836, 1.3683, 1.4399, 1.8458, 1.4770, 2.1799, 1.8113,
1.7162, 1.4880, 1.9263, 1.8749, 2.1199, 1.5113, 1.1552,
1.7085, 1.8008, 2.0422, 1.8567, 1.6550, 1.6946, 2.6835,
1.7319, 1.3809, 2.8460, 1.7601, 1.8997, 1.8239, 2.1551,
1.6634, 2.1943, 1.8340, 2.0278, 1.3254, 1.2468, 2.3133,
1.5443, 1.7584, 1.3926, 1.4852, 1.7444, 1.9240, 1.7073,
1.2002], device='cuda:0')), ('layer4.2.conv3.weight', tensor([[[[ 3.7087e-03]],
[[ 5.6045e-03]],
[[-2.3754e-02]],
...,
[[ 8.3045e-03]],
[[-2.0182e-02]],
[[ 8.7637e-03]]],
[[[ 2.1807e-03]],
[[-3.8605e-03]],
[[-1.4953e-03]],
...,
[[-8.2157e-03]],
[[ 2.4847e-02]],
[[ 5.1828e-03]]],
[[[ 6.2930e-03]],
[[-7.7607e-03]],
[[ 1.6016e-02]],
...,
[[-1.6587e-03]],
[[-2.0809e-02]],
[[-7.3934e-03]]],
...,
[[[ 2.8057e-03]],
[[-7.8960e-03]],
[[ 1.0590e-02]],
...,
[[-5.9861e-03]],
[[-1.0939e-02]],
[[-1.0955e-02]]],
[[[ 4.6868e-02]],
[[-2.1306e-02]],
[[-2.0913e-03]],
...,
[[ 6.0113e-03]],
[[ 1.0733e-02]],
[[-3.1923e-03]]],
[[[-3.1904e-04]],
[[ 1.7694e-03]],
[[-5.1331e-03]],
...,
[[-7.3763e-03]],
[[ 1.5524e-02]],
[[-1.0022e-02]]]], device='cuda:0')), ('layer4.2.bn3.weight', tensor([ 0.6621, 0.8399, 0.7994, ..., 0.5708, 0.7939, 0.5732], device='cuda:0')), ('layer4.2.bn3.bias', tensor([ 0.0179, 0.0589, 0.0418, ..., 0.0260, 0.0273, 0.0122], device='cuda:0')), ('layer4.2.bn3.running_mean', tensor(1.00000e-02 *
[-1.0790, -1.4523, -0.1461, ..., -0.4256, 0.7691, -0.4030], device='cuda:0')), ('layer4.2.bn3.running_var', tensor(1.00000e-03 *
[ 0.9161, 1.2459, 1.2937, ..., 0.4759, 0.9558, 0.5765], device='cuda:0')), ('fc.0.weight', tensor([[ 2.1340e-03, 1.3437e-02, 5.9206e-03, ..., 6.5675e-03,
1.2001e-02, 9.0808e-03],
[ 5.8618e-03, -2.1759e-05, 2.0472e-02, ..., 2.0673e-02,
1.1214e-02, -2.4791e-02],
[-7.2001e-03, 1.9593e-02, -1.4685e-02, ..., -2.2089e-02,
-2.6362e-03, 1.5685e-02],
...,
[ 1.7068e-02, -2.7486e-03, -2.9920e-02, ..., 7.7138e-03,
-1.8549e-02, -1.6591e-02],
[ 7.2249e-03, 2.5909e-03, -1.4177e-02, ..., -8.2361e-03,
-1.6696e-02, -1.4425e-02],
[-1.3319e-02, -1.8495e-02, -2.3692e-02, ..., 4.1775e-03,
-3.3484e-04, 2.2477e-03]], device='cuda:0')), ('fc.0.bias', tensor(1.00000e-02 *
[ 0.0113, -0.2409, 1.9775, 1.3757, -0.6795, -0.5739, 1.4860,
1.5876, -0.6300, -1.9958, -0.6843, 1.4052, 0.8061, 1.5009,
-0.9179, -1.7484, 1.2712, -1.1587, -2.0208, 0.9980, -0.5465,
0.5426, -1.6699, -0.0963, 1.5619, 1.9425, 0.5869, 2.1509,
-0.1626, -0.1914, -1.2073, -0.9013, -1.7466, -1.0571, -1.1594,
-0.2022, 0.9293, 1.4522, -0.8342, 2.3116, -0.1162, 1.7262,
-0.8329, -1.5757, -0.6641, 1.6176, -0.0551, 2.3419, -1.4863,
1.2244, -0.2306, -0.3516, -1.7553, 0.9367, -1.6599, -0.9016,
0.4911, -0.3354, 0.7276, 0.0984, -2.0717, 1.4959, 2.0143,
2.5933, -1.3597, 2.3482, 1.3203, 0.2829, -1.5683, -0.6384,
-0.5005, -0.7119, -0.0309, 2.6227, -1.2905, -1.4519, -1.7101,
-0.4961, -0.4697, 1.2190, 0.5659, -0.5647, -0.7892, 1.8892,
0.7559, 0.7938, 1.0923, 1.1040, 2.6188, -0.8994, -0.0477,
1.8627, 2.6044, -1.1527, -1.9355, 0.0554, 0.9116, -1.6275,
-1.6557, 1.0935, 0.3619, 0.7928, 1.0876, 2.0360, 2.1328,
2.3228, 0.0106, 0.7956, 1.5922, 0.0262, -1.4371, -1.2316,
0.2207, -1.1086, 0.4933, -1.2438, -0.0258, 2.4167, 1.2293,
0.7428, -0.3010, -1.7076, -1.3319, 2.0819, -1.2189, 0.3054,
-0.7012, 2.0958, -0.6215, -1.2097, -1.6367, 1.0193, -0.9783,
0.6253, -0.0256, -1.5992, -0.5596, 2.4556, -0.5759, -0.5137,
2.3781, 0.3374, 2.3880, 0.6803, 2.3074, 2.7585, 0.4683,
-1.1194, 1.9988, 1.6147, -1.7685, -1.1868, -1.5383, 0.8782,
-1.4547, 0.1162, -0.5350, 0.1194, -1.2702, -0.1574, -0.8426,
2.3082, -0.6150, 1.5548, 1.5189, -1.3246, 0.3603, 1.5595,
-0.0998, -1.4735, -0.3430, 1.3668, 0.6757, 0.7193, -0.0617,
-2.1342, 1.8586, -0.6090, 1.6667, 0.8360, 0.4612, 0.1989,
0.9398, 1.0539, -0.8777, -1.4809, 0.0704, -1.5489, 1.3089,
-1.5038, 1.3742, -0.5128, -0.7320, -1.2291, -0.9072, -1.4255,
0.7758, 0.7050, -1.5259, -0.3458, -1.6610, 1.6160, 0.6868,
1.3623, 1.5344, -1.1079, -1.5779, -1.3178, -1.1609, -1.1933,
-0.3213, -0.1677, 1.8325, 1.6540, -1.6659, 2.2937, 2.3721,
1.1145, 2.0883, 1.5251, -0.8382, -1.6707, 1.2584, 2.3694,
0.2262, -0.5856, -1.0443, -0.5088, -1.8725, 2.0324, -1.7196,
1.8816, 0.1075, 0.9588, -1.7850, 1.5972, 1.4570, -1.7243,
-0.1403, -1.1678, -1.4864, 0.0374, -1.3346, -1.6470, -0.6729,
-0.8838, 2.5543, 1.9080, 2.0796, 1.4260, -1.1667, -0.9568,
-1.8075, 1.9718, 2.1847, -0.0365, 0.4876, 2.5647, 2.2931,
1.3675, 2.5226, -0.6480, -1.1020, 0.6904, -1.3104, -0.2924,
1.9580, 0.5749, 0.9066, -0.4318, 1.5885, -0.2609, 0.6019,
1.2844, -0.7125, 2.0302, -2.0758, 0.0622, -1.8764, 2.3485,
-0.2963, 2.1007, -1.3858, 1.0481, -1.1103, 0.6025, -0.1805,
0.7161, -1.4811, 0.7359, -0.2307, -0.7759, -1.6213, 0.2537,
0.5142, 0.1664, 1.9516, 2.1663, -0.5716, 1.4361, -1.2296,
-0.8086, 1.8437, 1.7235, -1.3851, -1.5398, 0.8475, -0.9346,
1.7220, 0.3541, -0.3308, -1.0582, -2.0528, -1.4274, 1.7401,
-1.0712, -0.0534, 0.7450, 1.7922, 0.9021, 0.2933, -0.3090,
-0.6935, -0.9245, -1.9005, 2.2922, -0.4625, 1.6937, 1.5269,
-0.0864, 0.5110, 0.8392, 1.7447, 0.5624, 0.7616, -0.7311,
-1.9255, 0.4614, -1.0917, 2.4450, 2.1526, -1.0122, 0.3353,
0.9957, -1.4050, -0.5466, 0.2217, 0.6316, 2.2216, 0.9245,
-1.7310, 0.2007, -0.5298, 1.6959, -1.8161, 0.3242, -0.2712,
-0.4448, 1.7706, -1.8407, -0.6305, -1.9437, -1.1341, -1.0433,
-1.5199, 2.4443, 1.4083, -1.0599, 1.7801, 2.3451, 0.2489,
1.1294, 1.6025, 0.6153, 0.8642, 0.6096, 0.7351, -0.7135,
-0.9814, 0.3137, -0.3289, 1.8867, -1.7492, 2.1043, 1.1642,
0.7522, 2.3516, 1.3483, 1.8775, 1.6509, 0.3245, -1.4769,
-1.6485, 0.2761, -0.3868, -0.3236, 0.2349, 0.8465, 0.5585,
-0.1729, 2.3503, -1.2263, 1.3408, 2.2576, -0.5769, 1.3512,
0.8592, 1.5340, -0.2669, 2.1414, -1.1079, -1.6236, 2.1612,
-1.4161, 0.1078, -1.9567, -1.5410, 0.3847, 1.5820, 0.1216,
0.2918, -0.0423, 2.5674, 1.4908, -0.9168, -0.3830, -0.7193,
-1.1442, 0.0727, -2.1242, -1.5114, -0.6689, -1.2322, -0.6416,
-0.3858, 2.1682, 2.4615, -1.1352, 0.8928, 0.7486, -1.3124,
-0.3396, 1.7657, -0.1066, -1.4064, 2.2907, 2.3014, -0.0936,
-0.2298, 0.4698, 2.3842, -1.3449, 0.2783, 2.4419, 0.4964,
1.5834, 2.1826, 2.2147, 1.5345, 2.3620, -0.8531, 1.9917,
-0.4942, 1.2857, 0.7600, -1.9896, 1.6791, -1.2073, -1.8713,
-1.5707, -1.8106, 1.4034, -1.5195, 0.2528, -1.5925, 1.5287,
0.3886, -0.1498, 2.3039, 2.0583, 0.6231, -1.6926, 1.7134,
0.0542, 0.2619, -0.6407, 0.3559, -1.0554, 1.0394, -1.4076,
-1.9311, 0.4866, -0.1510, 1.1359, 1.4657, -0.0587, 1.9393,
-0.6344, -1.1554, -1.9270, 1.4553, 2.2878, 1.7257, 1.5867,
-1.3031, -1.1124, 1.7400, 1.7192, 1.4040, 1.4486, 2.6032,
0.9045, 0.3282, 1.7316, -0.8352, -0.4195], device='cuda:0')), ('fc.2.weight', tensor([[ 1.5058e-02, -1.9733e-02, 4.4920e-02, ..., -1.4017e-02,
7.6721e-02, -6.5122e-02],
[-3.7975e-02, 8.5738e-02, -4.6744e-02, ..., 2.7455e-02,
-4.0371e-02, 8.5722e-02],
[-3.6928e-02, -2.1672e-02, -3.6939e-03, ..., 1.7743e-02,
-4.3768e-03, 8.7593e-02],
...,
[ 1.2045e-02, -5.1611e-02, 2.6587e-02, ..., 4.1274e-02,
1.9180e-02, 2.9567e-02],
[ 1.4629e-02, -8.9040e-03, 2.3623e-02, ..., 2.8746e-02,
-5.4197e-02, 4.3382e-02],
[-3.5362e-03, -4.9038e-02, -1.0458e-02, ..., 6.1059e-02,
8.9643e-02, 3.9517e-02]], device='cuda:0')), ('fc.2.bias', tensor(1.00000e-02 *
[ 2.8155, -2.9263, -2.7312, 3.0690, -0.1653, -4.0859, 1.9346,
-2.8054, 3.8386, -3.1065, -0.5908, -0.4947, -0.3626, 2.0920,
-2.9381, 0.0207, 4.5065, -2.4694, 4.2461, -0.9133, 0.2913,
-2.8175, -0.7207, -0.2757, -2.2356, -2.5802, 3.2791, 0.4826,
-2.6665, 2.4677, -2.2298, 2.3499, -0.6537, 3.6268, -1.6083,
4.4148, -3.4655, 0.1602, 3.6827, 2.4749, -1.6761, -3.9401,
-1.4051, -0.4139, -0.9097, 0.1860, 1.0526, 2.9222, 3.6607,
-2.1005, -2.1388, 3.5179, -1.6914, -3.1769, 3.6535, 1.4361,
3.1154, 4.6070, 2.4958, 3.5244, -3.5650, -5.0715, 1.2197,
2.7051, -3.7806, 2.6928, -0.7314, 3.5128, -4.1509, 0.1046,
2.5881, -4.2578, -1.1336, -0.6362, 0.2998, 2.3658, 1.2526,
-0.6222, -4.4641, -0.2286, 3.1243, -3.4798, -0.5748, 1.6651,
-0.8110, -1.9712, 2.5939, 1.0374, 0.0989, 2.8978, -2.3218,
1.6227, -2.6673, 1.9006, -0.8733, 2.1731, -2.4252, -1.6938,
0.7728, -0.6866, 2.7433, 3.5003, 3.3816, -2.5184, -3.4905,
0.0418, -0.8208, 0.4348, 0.1088, -0.5476, -4.2553, -2.2980,
-1.6675, -3.7371, 1.0036, -0.4929, -1.5909, -4.8101, -1.6036,
-1.4782, -0.9225, 2.1591, 2.3182, -2.8294, -3.1920, 3.5901,
-1.8076, -4.1903, -4.9659, -2.8236, -2.1268, 1.6480, -3.1915], device='cuda:0'))])
(IMPLEMENTATION) Test the Model¶
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.
In [105]:
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
Test Loss: 0.343228 Test Accuracy: 89% (747/836)
(IMPLEMENTATION) Predict Dog Breed with the Model¶
Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.
In [108]:
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.
# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = [item[4:].replace("_", " ") for item in train_data.classes]
print(class_names[0])
Affenpinscher
In [109]:
def predict_breed_transfer(img_path):
# load the image and return the predicted breed
transform = transforms.Compose([transforms.Resize(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406],[0.229, 0.224, 0.225])])
file = img_path
file = Image.open(file).convert('RGB')
img = transform(file).unsqueeze(0)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model_transfer.eval()
with torch.no_grad():
out = model_transfer(img.to(device))
ps = torch.exp(out)
top_p, top_class = ps.topk(1, dim=1)
index = top_class.item()
return class_names[index]
In [110]:
predict_breed_transfer(dog_files[100])
Out[110]:
'Doberman pinscher'
In [118]:
import matplotlib.pyplot as plt
import matplotlib
def display_breed(file):
file = Image.open(file).convert('RGB')
plt.imshow(file)
matplotlib.pyplot.text(5, -20, "HEllO, Lets Detect the BRE !",
color='black', fontsize=15)
plt.show()
Step 5: Write your Algorithm¶
Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,
- if a dog is detected in the image, return the predicted breed.
- if a human is detected in the image, return the resembling dog breed.
- if neither is detected in the image, provide output that indicates an error.
You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and human_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.
Some sample output for our algorithm is provided below, but feel free to design your own user experience!
(IMPLEMENTATION) Write your Algorithm¶
In [119]:
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.
def run_app(img_path):
## handle cases for a human face, dog, and neither
if (dog_detector(img_path)):
print("Dog Detected!")
breed_name = predict_breed_transfer(img_path)
display_breed(img_path)
print("Predicted breed is :", breed_name)
elif (face_detector(img_path)):
print("HUMAN Detected!")
breed_name_human = predict_breed_transfer(img_path)
display_breed(img_path)
print("Predicted closest breed is:", breed_name_human)
else:
print("Neither a Dog nor a Human :( :( !")
In [120]:
run_app(human_files[200])
HUMAN Detected!
Predicted closest breed is: Dogue de bordeaux
Step 6: Test Your Algorithm¶
In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?
(IMPLEMENTATION) Test Your Algorithm on Sample Images!¶
Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.
Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.
Answer: (Three possible points for improvement)
The output is not much good as expected as it is predicting sometimes wrong labels and this is due to network having not very high accuracy. Points for improvement :
We can try to introduce dropout or other regularization measures to avoid overfitting of model over the training dataset.
We can try more appropriate hyperparameter tuning methods such as grid search to fine tune the hyperparameters of the models.
We can check for gradient explosion problem by examining each layer's weights updates and analysing other metrics. Using Gradient clipping can help.
We can try to reduce batch size as it has been sometimes, high batch size leads to loss of generalization (maybe we can try 8, 16.. etc..).
We can also try to change weights initialization by using Xavier or He initialization because these can affect the learning process.
In [121]:
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.
## suggested code, below
for file in np.hstack((human_files[:3], dog_files[:3])):
run_app(file)
HUMAN Detected!
Predicted closest breed is: Chihuahua HUMAN Detected!
Predicted closest breed is: Australian shepherd HUMAN Detected!
Predicted closest breed is: Alaskan malamute Dog Detected!
Predicted breed is : Mastiff Dog Detected!
Predicted breed is : Mastiff Neither a Dog nor a Human :( :( !
MY testing¶
In [122]:
run_app('1.jpeg')
Dog Detected!
Predicted breed is : Maltese
In [ ]: